This is much colder than that, though not less human. This probable self lifts fingernails in the sense that at any given moment what exists socially, relationally, even epistemologically, is not a total self, but the most likely version of a self that can be assembled from the traces available under those conditions.

Not the true self as opposed to the false one, which is schoolyard philosophy for adults who still think authenticity is a stable mineral buried under enough excavation, but a weighted field of possible personhoods, some more latent, some more recurrent, some made suddenly vivid by pressure or context or memory, some rendered actionable because another person, or a platform, or an institution, or an archive, or an AI system, has enough pattern to proceed as though it knows you.

Here is the piece. Enjoy.

Baudrillard has never quite landed for me. I suspect part of the reason is that his work has so often arrived as a kind of secular prophecy, and having not experienced a religious upbringing, I have no mouse-trap to trigger there.

However, I know enough about a sermon to feel that Baudrillard’s work feels endlessly recycled through the same exhausted metaphors of disappearance, unreality, simulation, copies without originals, as though the deepest crisis of modern life were that the image had somehow broken away from the thing. Even in the eighties, I think there was already something old in that mode of thought, something still too tethered to the historical weight of prior philosophy, too willing to cast itself as a revelation of loss, too seduced by the drama of exploding the ‘discourse’.

This is perhaps why so much popular engagement with Baudrillard ends up feeling biblical to me, or at least biblical by way of Hollywood, which is to say apocalyptic in tone, obsessed with fallen worlds and degraded originals, all of it perfectly suited to the flattened pseudo-metaphysics of The Matrix, where the problem can still be narrated as a break between the real and the false. There is the body and there is the digital realities of code, the true world and its counterfeit.

For me what seems far more materially obvious now, is not that reality has been replaced by signs, but that personhood itself has become increasingly inferential, increasingly assembled through partial traces, increasingly lived as something neither wholly present nor wholly absent. Now, a self is rendered into workable form by whoever or whatever needs enough of you to proceed.

Before and around Simulacra etc., cybernetics, systems thinking, computation, and information theory had already begun to reconfigure the human as something increasingly endangered by representation, and something now entangled with feedback, mediation, pattern recognition, storage, transmission, and adaptation. It always had been, but digital realities had created new channels, shapes, velocities, that could not be ignored.

The digital environment does not create this condition from nothing, and I think that is where so much writing on digital identity goes limp. So much prose seems dedicated to pretending that multiplicity itself is new, or at least looks unrecognisable enough now it’s worth dissecting, or that fluidity arrived with the interface, not applicable to the self before. But humans have always encountered each other through probabilities, through reputation, gesture, class signals, narrative residue, social positioning, remembered fragments, gossip, old injuries, bodily cues, institutional records, and countless other forms of partial access. No one has ever had the whole of another person.

The problem seems now that multiplicity is still apparent, but now it is indexed, stored, queried, resurfaced, cross-referenced, and operationalised at a scale and persistence that no longer allows partiality to remain partial in quite the same way. Fluidity, in other words, is no longer merely fluid, like water, now a self can be frozen, become porous, and importantly, gathered into vessels, drank, contaminated, run off your skin and down the drain.

Now, a self can be made to outlive the conditions that produced it. Rain, gas, snow, gone. Weather patterns too are kin, a self can be treated as predictive. Our self can be sold back to systems as certainty. We say 30 degrees today and chance of rain in the evening, and even if it doesn’t happen, we know we did something to assemble enough truth to keep providing the report the next day, minute to minute, or look up at the sky ourselves.

A “fluid self” has always sounded too romantic to me, too soft around the edges, too flattering in its implication of freedom or endless becoming. It avoids the contemporary stakes applied to self, we have moved past mutability and into the realm of addressability. Selves shift, but the fact that those shifts leave behind trails dense enough for other entities to build provisional models and then act on those models as though they were you, is where we exist now.

This, to me, is where reality and identity begin to blur, though not in the stupid sense that reality is somehow just subjective, or that the external world disappears into vibes, but in the sense that social reality is increasingly composed through inferential coordination, through overlapping systems of approximation that decide what version of a person is real enough to matter in a given moment.

A boss does not know you, exactly, but knows enough to decide whether you seem reliable.

A lover does not know you, exactly, but knows enough to risk disclosure or withhold it.

A government does not know you, exactly, but knows enough to classify you.

A platform does not know you, exactly, but knows enough to sort you into likelihoods and reconfigure the world around those estimates.

An AI does not know you, exactly, but knows enough, or seems to, to generate a response pitched to your rhythm, your tolerance, your obsessions, your recurring syntax of concern.

We even have a term for the peculiar despair that accompanies logging off: digital suicide. The fact that disappearance from the retrievable field is now felt through the language of death says more than people realise about the scale of the shift already underway. It is not addiction, dependency, narcissism, or the sad little moral panic people like to drape over contemporary life, but a more fundamental ontological disturbance, the creeping sense that to become unreachable, unsummonable, unqueryable, and that is to risk a kind of social and psychic unreality.

For some younger cohorts, particularly Gen Z and what is emerging in Alpha, this condition does not arrive as a corruption of some older fantasy of stable personhood, that is something they never had. They do not meaningfully inherit that fantasy from their parents, instead they come of age inside an environment where the self is assembled under archival pressure from the beginning.

Their environment is one where traces are expected to persist, where expression is ambiently cross-contextual, where fragments travel further than intention, and where the developing person learns, often before they have language for it, that what can be retrieved can be acted upon, and what cannot be retrieved begins to feel as though it risks not fully existing at all. To “be” and to form, they generate autonomy by choosing where to digitally locate themselves, and digitally present themselves, what happens in a physical reality, is simply a bodily expression of the digital environment they are constructed within.

Under these conditions, adolescence itself changes shape, because experimentation no longer disappears cleanly into time, phases do not dissolve with the privacy of passing through them, mistakes do not always decay into memory, and retreat does not necessarily offer refuge. A self is under a different kind of threat: the fear of falling out of circulation, of becoming harder to index, harder to prove, harder to stabilise to oneself and others.

At the same time, what does persist may harden too early. A joke is never only local. A face is never only a face. A preference, affiliation, tone, aesthetic, opinion, vulnerability, or mood may return later under altered conditions of scale and consequence, recoded as evidence, stripped of atmosphere, made to stand in for a whole person before that person has even had the chance to become otherwise. The result is not simply that younger people are more performative, which is the lazy diagnosis of adults who mistake environmental adaptation for moral decline, but that they are being formed inside a double bind in which selfhood is pressured from both directions at once.

On one side, the terror of disappearance, of no longer being retrievable enough to remain socially real; on the other, the terror of premature fixity, of being frozen by residue before one’s latent selves have had the conditions under which to emerge. One grows up, then, not only by becoming, but by being shadowed in advance by the versions of oneself that systems, peers, schools, platforms, institutions, and archives may later be able to summon, and by learning that existence is no longer experienced solely as embodiment or relation, but increasingly as the management of one’s position within a field of possible retrieval.

In this world, to vanish feels like a kind of death, and to remain too available risks becoming a prison before a self has even finished forming.

A self however you spin it, is not unreal. But it is assembled under conditions of incompleteness, out of patterned information, and made consequential before fullness is ever possible.

That is why I find AI clarifying here, because it reveals the mechanics with less sentimentality. Every time I write a prompt and receive a response, what returns to me is shaped not only by what I have said in that instant but by the model’s probabilistic construction of who I am likely to be, what I am likely to mean, what register I am likely to want, what conceptual density I am likely to tolerate, what irritates me, what interests me, what counts, in this context, as enough coherence to proceed.

It is not reading my soul, obviously. it is assembling a working self out of residue. But another person cannot read my soul either.

Whether you like to admit it or not, but a human as much as an AI does not invent inferential personhood so much as they expose how much of ordinary interaction already depends on models of likely selves being built from incomplete traces and then mistaken, after the fact, for intuition, chemistry, judgement, common sense, or understanding. The machine strips the romance off what has long been true: that much of social life consists in compressing one another into sufficiently workable forms.

To call this a probable self is not to deny embodiment, nor to claim that everything is data now, nor to celebrate the collapse of coherence into some fashionable postmodern fog (again, sorry Baudrillard). It is to say, more narrowly and more precisely, that selfhood may be better understood not as a fixed thing expressed through different channels, but as a weighted distribution of possible actualisations, made more or less likely by context, pressure, memory, risk, desire, interface, and the conditions of retrieval. Weather.

Some versions of us recur strongly enough to be mistaken for essence. Some appear only under duress. Some are vivid to ourselves and invisible to others. Some are archived so heavily they remain active long after they have ceased to feel experientially alive. Some are barely lived at all, yet become socially real because they are the ones most easily indexed and acted upon.

Some selves are not absent, false, or suppressed, but latent. Meaning, they present as possibility without yet having acquired enough residue to become retrievable, legible, or socially consequential. Latency distorts our perception that everything real about a person arrives in a form that can be indexed on demand. But, some ways of being remain dormant until a context calls them forth, until a relationship permits them, until pressure forces them into surface, until time itself renders them available.

The violence of probabilistic identity systems is not only that they misread what is there, but that they collapse latency into premature certainty, treating the currently retrievable self as though it were the whole field. A person is then not merely simplified, but frozen before they have had the conditions under which other versions of themselves might emerge.

If older philosophy was obsessed with what makes a person the same over time, perhaps what matters now is the inverse question. So, if what the digital environment has changed is not merely how we present ourselves, but how our possible selves are stored, made probable, and converted into reality by entities that do not need to know us fully in order to shape the worlds we inhabit, what can be discarded in old conclusions or conceptualisations around self-hood, and what can be generated in their place?

Memory is central to this, because memory is so often treated as the guarantor of selfhood, as though to remember were to preserve, and to preserve were to maintain some meaningful continuity. But in practice memory is far less stable, far less noble, and far more selective than that. Memory is not a vault, private in one way, public in another, it acts as a weighting mechanism. Memory is a way of increasing the future availability of certain versions of a person over others.

What memory stores follows patterns of retrieval, in code and in person, both are ruled by a biased architecture of access, where certain gestures, failures, moods, phrases, injuries, performances, and disclosures become disproportionately easy to call back into the present, while others erode, drift, or are never encoded strongly enough to remain socially legible at all. To remember someone, then, is already to produce a probable self, because what returns is not the whole person but the version of them made most retrievable by affect, repetition, narrative usefulness, trauma, love, resentment, institutional format, or simple cognitive economy. Memory does not merely hold the self in place. It distorts the field of what that self is likely to become again, at least in the eyes of whoever is remembering.

This is why the digital archive matters so much, and why I think most discourse about online identity remains strangely shallow even when it gestures toward surveillance or permanence. It stops too often at exposure, as though the problem were simply that too much is visible, but really our discomfort is that our visibility can be retrieved in any form, at anytime, without us having to truly ‘exist’. Searchability is not identical to seeing. An archive is not merely a storage container. It is an engine of differential return.

Once traces are indexed, certain selves become easier to summon than others, easier to link to one another, easier to frame as continuity, easier to operationalise as evidence. The old post, the old image, the old phrasing, the old preference, the old association, all of it can be made to stand in for a person’s present reality not because it is truer than what is currently lived but because it is more accessible to the systems and observers now acting upon them. What survives is not always what matters most. It is often merely what was formatted well enough to persist.

So much language around identity and selfhood remains clogged by moralism, by authenticity discourse, by the dead-end demand to separate what is real from what is fake, as though the key problem of modern life was still sorting out what was true and what was false. It cannot be proved, we feel it as we argue, for the lived self is the self as it exists in experience: messy, inconsistent, often opaque even to itself, full of intensities and absences that leave little residue.

The inferred self is the self assembled by another person, or another system, from the traces available to them, whether those traces are linguistic, behavioural, aesthetic, bureaucratic, historical, or computational. The actionable self is the version of the person that actually enters into consequence, the version on which a decision gets made, a judgment formed, a door opened or closed, a response generated, a recommendation surfaced, a punishment applied, a relationship altered.

These three do not align neatly, and perhaps modern identity crisis is the conflict between them, about the increasing distance between the self one inhabits, the self others infer, and the self that systems are prepared to act upon.

It is cheap too to reduce the source of this to empty or overused categories, like capitalism. Naming an economic totality does not solve the conceptual problem. Specifically, capitalism matters here, yes, but now it is as amplifier, allocator, and extractor. It does not invent probabilistic personhood, but it does industrialise those models, reward their circulation, and tether them to systems of value in ways that make certain inferences difficult to escape. When legibility is monetised, it turns traits into segments, tendencies into markets, rhythms into engagement loops, habits into risk profiles, expression into metadata, and the uncertainties of social interpretation into increasingly hard infrastructures of sorting and response.

But to stop there would be to make the same mistake I keep resisting, which is to let capitalism function as the final explanatory brick wall, the place an essay walks into when it needs closure. The deeper issue is older and wider than that. It is the fact that any world composed of partial observers must operate by approximation, and that approximation always produces a version of the self sufficient for action before truth has been exhausted.

We can talk forever about selves as narratives, masks, performances, multiplicities, but what turns theory into structure is the question of consequence. In reality, any reality, which version of a person is treated as real enough to move the world around them?

A person may feel transformed, and yet remain actionable to others only as an old version of themselves. A person may be misread entirely and still have that misreading become the basis of institutional reality. A person may be known intimately by those around them and yet reduced by digital systems to a thinner, more stable, more predictive profile that governs what opportunities, risks, prices, or forms of attention become available to them. In that sense, the probable self is not simply the self as guessed. It is the self as guessed under conditions where the guess can do something.

Every system operates with its own threshold for acting on those inferences and its own tolerance for uncertainty before approximation hardens into consequence. A friend needs very little to begin relating to you as changed. A lover may act on tone alone. An employer may require a pattern. A hospital may require a number, an image, a documented event. A platform needs almost nothing at all, only enough behavioural confidence to sort, recommend, suppress, or intensify.

The actionable self is therefore never produced at one universal level of truth, but at shifting thresholds of action, where different entities decide, implicitly or structurally, how much of a person they need to know before they are willing to proceed as though they know them. This is where the probable self becomes hard to hold, because the problem is not simply that we are guessed at, but that the threshold for treating the guess as real is often far lower than the threshold for correcting it.

There is no stable person waiting to be uncovered, expressed properly, aligned with itself, recognised at last. That fantasy increasingly feels like faux-nostalgia used as crutch for a mode of personhood that may always have been administrative rather than real. The stable self was never simply discovered, it was demanded. It was useful to law, to labour, to biography, to kinship, to morality, to any system that required continuity in order to assign responsibility, trust, debt, inheritance, guilt, or value.

What digital conditions have done is not abolish that demand but expose its fragility by multiplying the contexts in which a person can be rendered differently and by storing enough traces for those renderings to be compared, replayed, and turned against one another. The probable self therefore does not describe a new fragmentation but a newly externalised condition: the fact that selfhood has always been assembled under uncertainty, but that uncertainty now leaves behind indexed residue.

A residue has a legacy, a particulate afterlife of a person moving through systems, a compression left behind when a life is too large to be held. These must be read as samples, ones that can be searched, deleted, replaced, edited. That’s the construction of new self.

From a prompt history. A performance review. A search result. A rumour. A username. A credit score. A tone. A sentence someone cannot forget. A photograph that keeps returning. A way of responding that becomes expectation. A model does not need a soul to proceed. Neither does an institution. Neither, often, does another human being. All require sufficient residue to build a probable self and act accordingly. So much of what we call understanding may always have depended on this, which is not easy for the mind to comprehend.

I think we need to insist that our theories of self need to become better at dealing with conditional actuality. The self is not therefore unreal, but nor is it best conceived as a singular object moving cleanly through time. And if that is right, then the philosophical task is no longer simply to ask who one really is, but to ask how selves become probable, how probabilities become actionable, and how action, once taken, feeds back into the very reality it first only approximated.

What this displaces, then, is not merely one philosopher or one stale digital cliché but an entire moral architecture built around the fantasy that personhood is best understood through coherence.

There is no such thing as distortion, as being in two places at once, because lies exist only as long as they are proven, and a stable self needs proof only as an administrative requirement. Stable selves are convenient to institutions that need people to remain narratively continuous enough to be tracked, judged, hired, disciplined, trusted, inherited from, married, blamed, redeemed, diagnosed, and governed. The cohesive subject is not a metaphysical truth that modern technology has tragically broken, it is a bureaucratic ideal, a social convenience, a legal fiction with extraordinary staying power.

The deeper issue is that these partial renderings do not remain where they were produced. They travel. They sediment. They are remembered unevenly, stored badly, weighted selectively, linked retroactively, and increasingly made available to systems that do not merely observe but sort, predict, and intervene. A self is therefore not simply presented. It is circulated as residue. It is compressed into enoughness.

The lived self is not truer because it is private, nor is the inferred self false because it is partial; partiality is the condition under which almost all social knowledge takes place. But action introduces an asymmetry that philosophy too often evades. I may inhabit one version of myself and be acted upon as another. I may have changed in every experiential sense available to me and still remain institutionally legible only through stale residue. I may be read more accurately by a friend than by a platform, but the platform’s poorer reading may still have greater material consequence. I may be fully known in love and badly known in law. I may understand myself with extraordinary precision and still fail to become actionable in the world except through thinner, uglier, more searchable approximations. That, to me, is where contemporary identity feels unstable: because different systems act on different probability models of the same person, and those models do not share a common scale of truth, intimacy, or consequence.

An archive must be treated not as a passive backdrop to selfhood but as one of its active conditions. An archive does not only remember; it privileges, and what it privileges begins to feel like continuity. The old image, the cached line, the résumé, the diagnosis, the search result, the school report, the purchase history, the chat log, the anecdote retold too often, all of these do more than preserve a life. They weight it. They pull future interpretations toward some versions of a person and away from others. They give institutions and observers shortcuts, and shortcuts have always been one of the hidden motors of social reality, not because they are accurate but because they are efficient enough to act on.

Again, this is why so much talk of visibility misses the point. The issue is not simply that we are seen too much, though we are, nor that information persists too long, though it does. It is that certain traces become infrastructurally available in ways that allow one version of a person to harden into the default referent against which everything else must struggle to count as real. The archive, then, is not the enemy because it contains too much truth, but because it contains enough residue to counterfeit continuity.

The danger, of course, would be to let this harden into a total theory and begin speaking as though the self were nothing but a dataset in motion, a Bayesian ghost, an optimisation problem with literary ambitions. That would be just another flattening, and a stupid one. The probable self is not meant to abolish embodiment, contradiction, opacity, or the irreducible excess of lived experience.

If anything, it only becomes necessary because persons exceed every model built of them. But excess does not protect us from consequence. That is the point. The fact that a person is more than the residue through which they are known does not prevent that residue from governing what happens to them. A life can overflow every available category and still be cornered by the categories most actionable in a given system.

This is why the question of selfhood can no longer be treated as merely expressive or introspective. It is infrastructural, it has thresholds.

How much certainty does a system require before it acts as though it knows who you are? How much residue does a person have to leave behind before they become narratively stable to others? How much misrecognition can a life absorb before it begins to feel more real, socially speaking, than the life actually being lived?

This is not a world of neutral interpretation. This new version of self shapes the opportunities through which a person may next become legible, the risks they are made to carry, the forms of explanation they are forced to provide, the kind of future they are permitted to have. A person who is repeatedly inferred as unstable may find stability made less available to them. A person rendered as desirable may be trapped by the very coherence of that desirability. A person archived as deviant, difficult, exceptional, damaged, gifted, trustworthy, dangerous, promising, or incoherent may find that the probability model built around them begins to pre-structure the reality from which later interpretations are drawn. In that sense, the probable self is not just an epistemic object but a recursive one: action based on approximation helps produce the conditions that make the approximation appear retrospectively justified.

So no, Baudrillard, the philosophical task is not to mourn the loss of a stable function that may never have existed outside institutional convenience, nor to celebrate fluidity as though mutability were inherently liberatory, but to understand how signs and selves become sufficiently real to matter under conditions of partial access, and how those conditions are now organised by systems of storage, retrieval, memory, and prediction that operate with extraordinary confidence despite their constitutive incompleteness.

The question is no longer who I am, full stop, nor even who I seem to be, but which version of me acquires more force in the world than the one I am currently inhabiting. If older philosophy asked what makes a person the same across time, now we must ask what makes one rendering of a person durable enough to become their reality in the eyes of others, and what forms of life become possible or impossible once social existence is increasingly organised not around essence, but around indexed likelihood that they are here enough to matter to reality, but not necessarily to themselves.

I got a pop-up the other day whilst catching the train, and ended up watching a livestream conversation titled: AI and the Future of Media. The chat was between The Atlantic CEO Nicholas Thompson and tech thinker Azeem Azhar, hosted on Azeem’s Substack, Exponential View.

It's here.

The section I jumped into was midway (I think) and Thompson was describing how AI models scrape and summarise journalistic work, including The Atlantic’s, without any compensation. GPT-style systems ingest paywalled prose and regurgitate it in free, stitched-together (coherent enough…) answers.

This causes traffic drops, ad revenue collapses and subscriptions stall. They were already borderline unviable, but now the not-quite-theft of journalistic labour isn’t fuelling innovation for anyone except OpenAI and their institutional kin. Even independent journo off-shoots are still chucking up paywalls on their content because they simply do not know another way to get paid.

Publishers are indeed sprinting in all directions. You’ve all been annoyed by a slapped up paywall, but behind the scenes everyone’s flirting with blocking crawlers (again, monopolised, not transferable, concentration of serious power) and drafting lawsuits on company letterhead.

In scary news, a few heavyweight incumbents (Axel Springer, the Associated Press, others with lawyers on speed-dial) have already inked licensing pacts with OpenAI, hoping to swap access for survival. The uniformity is not a thing, it’s both a stampede and side-door diplomacy.

Thompson compared this landslide / tiptoe to the early days of Google when SEO broke media business models until a new détente was brokered.

Reality is, even Google is panicking.

Gemini is a case and point, and due to Google’s configurations and well, knowing where the money is at, they are pushing AI summaries up top. If it keeps users on their page or clicking through to Gemini for answers, that’s good for them, but not for the actual sources it stole from below. Pew’s July 2025 study shows click-through-rate plunges from 15 % to 8 % when an AI Overview appears. This guts all traditional income for search engines and businesses, primarily ad revenue and subscriptions.

Publishing has shown their Achilles heel with paywalls and subscription models being the only way to remain viable, and well, pick up a newspaper and see how many ads you have to flick through before you get to a 200 word snippet that’s clearly got motive.

With Google a la old days, the threat and the tool were centralised, indexed, and crucially, negotiable. AI doesn’t work that way. The stakes, methods, and motives are different. The models are widely cloned and when someonthing is crawled and put into the system, it disassembles, becoming pure tokens. This idea filters to tech libertarians who start to use language like “decentralised” and lets AI off the hook—because they already swallowed everything, saturating the corpus. What’s “miasma” cannot be regulated.

(Yes, transformer weights store distributed statistical correlations, not literal passages. A ten-word qualifier for all the “well-actually” coming my way. I just wanted to make it jargon-lite, sue me)

So no, it’s not a traffic leak. The crawlers scrape, the models chew, and what comes out isn’t journalism (well not anymore ) and looks much more like minestrone. Not a finely crafted article, just tokens in the slot machine. AI companies know this.

We don’t really know why memory sometimes returns to us, much as consciousness still eludes us. Smell gives us an idea, but the main thing is that if it can be retrieved, it matters. And all these models do is retrieve, predict, spit out.

They say, “Hey, the piece is soup. It’s not verbatim.” But memory doesn’t work like that in humans or machines.

At the end of the day, the proposed solutions for publishers are cosmetic, contradictory, or philosophically unserious. Thompson rightly sees through it. Azeem too, though I suspect he’s hesitant to wade into a space this foggy, this speculative, this miasma.

Fair.

“Just block the bots” is wishful, but at least it contains hope. “Just sue” is slower and far more expensive than the model’s next release cycle. If I were big-tech, I’d budget for lawsuits the way Woolworths budgets for shoplifting.

“Just license”? Sure. Try it. But that assumes enforcement is even possible. The internet is still the wild west; Azeem mentioned that China now files more AI patents than any industry in history. What the hell is everyone building? We’ll soon see, but one thing is clear that this landscape doesn’t bend to licensing. It eats licensing for breakfast.

“Just…” is a battle cry resulting in echo-chamber solutions. The posture from tech is all too familiar: band-aid, bullet point, vague promise to circle back in Q3.

Whilst they stall, the media is treating it like a sprint, but half the runners don’t know which way the track curves. Legacy industry leaders are already bleeding out on the bitumen and academia is further gone—beyond roadkill, now one with the open road.

OpenAI’s Media Manager is the most visible attempt to throw a leash on this problem, and it’s telling that even that’s gone nowhere.

Announced in May 2024, it was meant to be a tool that lets creators and publishers flag their work to be excluded from future training runs. There was no detail on how it would work, no enforcement mechanism, no audit layer. Just the usual choreography: blog post, press pickup, breathless LinkedIn threads. As of writing, it’s July 2025 and still nothing has shipped.

The project lead quietly moved to part-time consulting months ago, and even the trade press seems to have clocked it as vaporware. But beyond the missed deadline and the suspicious silence, the real problem is this: Media Manager was never designed to remove content from trained models, it’s just a robot.txt with a fancier haircut.

What I mean by that, is that from what I can ascertain, it doesn’t touch what’s already embedded in the weights. Meaning there’s no unlearning, no gradient intervention and certainly no pathway for a publisher to say: this specific article, this specific phrase, this specific pressure on your model—undo it. It is all stall, and I get it OpenAI, when tech moves as fast as it does, stalling is the major weapon.

What people seem to be missing about Media Manager, is that it assumes that trust is the currency: you submit your content and hope OpenAI honours the flag. It’s not the same though; Nicholas and Azeem talk about conviction as the currency of digital publishing, but there is zero conviction in the tech industry and it has given up on the idea of trust, it works more on the currency of social amnesia and momentum.

What this means for publishers is they’ll be stuck hoping they don’t move the goalposts later. Hoping that your work wasn’t already in GPT-3.5, GPT-4, or whatever internal Frankenstein’s monster they’re fine-tuning this week.

That’s not governance, that’s a hostage negotiation, and OpenAI is already giving the cops their pizza order.

The longer Media Manager stays frozen, the clearer it becomes that the tech wasn’t the blocker and the faster others can get to work. The implications are the issue of releasing such a tool to the public, because building a true forgetting system is to admit that memory in these models is erasable. That’s a whole big business deal breaker when your whole schtick is that AI does what 400 million brains can do by shoving all of the internet in it and sifting for you.

Once you admit that, you’re cemented on the hook and you are getting eaten for dinner. Across all factors too, ethically, technically, financially, and legally. So, Media Manager avoids that, and I’m sure delays it’s release to try and cushion and protect the model. But honestly, fuck those guys, I truly believe that’s why the next move can’t come from inside the house and needs a more proactive and less strings attached team on this particular job.

So, when “Just block the bots” is a fence post and useful only until the wolves learn to jump, and “Just license” is a treaty no one can police once the crawl has already happened, you need an actual procedure that can be rote-followed. One that can remove the splinter already buried in the model’s nervous system without amputating the limb and having to start all over.

I had a think about it, and re-visited an old bit of protocol I was working on after thinking about the failings of crypto (I spite learnt what it was here). I looked at it from the AI unlearning angle, then asked Claude Opus4 to pose the question I’m really looking at in detail, so I could attempt to solve it.

“Design a system that allows AI models to selectively "forget" specific training data after deployment while maintaining overall performance. The system must:

1. Identify and remove the influence of particular data points from an already-trained model

2. Prove that the targeted information has been effectively removed without revealing what was removed

3. Maintain model performance on unrelated tasks after removal

4. Scale efficiently across models with billions of parameters

5. Handle requests from multiple parties with potentially conflicting removal demands

6. Operate without retraining the entire model from scratch

7. Create an audit trail that proves compliance without exposing private information

8. Work across different model architectures (transformers, diffusion models, etc.)

This is a real unsolved problem in AI governance touching on privacy rights, copyright, and the "right to be forgotten."

Okay goodbye AI, time for me to think about this properly.

The reason frontier vendors avoid building real forgetting systems isn’t anything to do with society, it’s because it is expensive in all senses of the word (yes, that’s ALSO electricity I’m talking about). To scrub something from a model after training, you first need to know exactly how it got in and what part of the data pushed which neuron into what shape.

But during pre-training, influence bookkeeping doesn’t exist because that would be mathematically enormous to execute and the people running the show are not super-geniuses, they are employees.

Technically, nobody saves per-sample gradients or maps how a specific article rippled through the architecture, because that’s like going through every single atom in the space you’re blowing up to make a quarry. Tech wants more quarries, and they want them now.

Reconstructing everything you blew up in the first place, means simulating a second crawl. This requires terabytes of data pushed through the model again just to isolate one faint shadow of influence. To go back to my mine analogy, that’s like building a secondary “fake” mine to compare and contrast with your blown up one.

In computing, the amount of energy and memory this would consume is unfathomable, and probably would require people to ditch Australia as a place where anything is and instead make it one giant server farm. I’m sure this would happen by 2050 regardless, but if push comes to shove, this issue in AI ethics will make it come much sooner. You’d need petabytes of scratch I/O and burn weeks of compute doing it.

Ever heard of a petabyte? It’s big, and not as cute as it sounds.

Which means the moment you try, you’ve blown past the capital expenditure (the money an organisation or corporate entity spends to buy, maintain, or improve its fixed assets, such as buildings, vehicles, equipment, or land.) limits that made your margin charts look good in the first place.

What does a business need? Investors.

What do investors want? Numbers.

But, these numbers are ugly and a petabyte is particularly no-go for them.

Stalemate.

This is all before you get to the bigger risk: that a poorly-aimed memory wipe doesn’t just forget what you asked it to, but takes out neighbouring concepts with it.

Nuke Harry Potter and your model might forget the entire orphan-wizard archetype. This will not make the folks at Dungeon’s and Dragon’s happy, and someone’s gonna snap when their Level 19 Wizard no longer has a heart-warming backstory.

Put bluntly, we don’t want tomorrow’s chatbot forgetting how to form a basic sentence because someone at The Guardian filed a takedown on a piece about imperialism in YA fiction. For companies, this isn’t a functional threat, but one that is entirely reputational. Daddy needs money to eat up all the wizards for the most probable result in your output.

The headline:

“HARRY POTTER HAS GREEN EYES”

ChatGPT fails to understand lore as artfully constructed by the world's most famous TERF.”

JK’s aside, say you managed to extract it, after you’ve decided let’s give it a crack and burn through some more fuel that could power 1 trillion homes.

The result? You’ve removed the memory and somehow it hasn’t broken the bot, but what remains is the interaction between the writer and the surgeon and it has absolutely zero trust intact.

“Prove you have removed my piece like I asked you to.”

“Trust us.”

“No.”

What currently exists to do this are leak probes which are effectively prompt-based statistical guesswork. These are fine for academic papers, but they fall apart in court because the guessing is up to a jury, not the judge. Again, reputation, reputation.

What is increasingly required is a proof that is rock-solid and even Sisyphus can find a spot to put it on top of the mountain.

As much as it pains me to mention due to my hatred towards the entire occupation of this corner of tech, cryptographic proof is probably the strongest answer. It’s the only thing that could attempt to show a piece is not in the weights anymore, without revealing what the idea was. That’s not trivial, not even a little bit, and it’s precisely how crypto sticks around.

(Resentfully, I endure the insufferable…)

In crypto, zero-knowledge proofs (ZKPs) let one party prove they know something—or have done something—without revealing what it is.Think of it as saying, “I’ve walked through the maze,” while showing a sealed envelope with the exit coordinates inside, and everyone else can verify the envelope is legit without opening it.

The most common use in blockchain is to prove that a transaction is valid (e.g. no double-spending, no counterfeit coins) without exposing the transaction’s details. zk-SNARKs (Zero-Knowledge Succinct Non-Interactive Argument of Knowledge) are a specific kind of ZKP that compress this proof into something small and verifiable quickly, without requiring ongoing back-and-forth between the prover and verifier.

This is crucial in crypto because it lets systems be both private and auditable. You get the best of both worlds: trust without transparency. You don’t have to see someone’s balance or transaction history to know their transaction was actually sound. That’s how protocols like Zcash work, and increasingly how Layer 2 scaling solutions like zk-rollups compress hundreds of transactions into a single cryptographic proof that gets verified on Ethereum.

Now it is my duty to remind the reader that crypto proofs are theoretical integrity bound to practical slop. Do not trust them because they have a cool looking lock-box. It’s still run by GitHub and Discord and there’s no traceability or accountability within the whole premise. Decentralisation is just a fancy term for ghosting.

Moving on.

Apply the crypto logic to AI unlearning, and you get a theoretical solution to a core problem: how do you prove a model no longer remembers something without showing what that something was?

zk-SNARKs could be used to generate cryptographic proof that the weights were modified correctly—according to a pre-agreed forgetting protocol—and that the model’s outputs changed in a way consistent with that erasure. The verifier (say, a regulator, a publisher, or a steward) wouldn’t need to see the training data or the private internal logic of the model, only the proof. This makes deletion requests both enforceable and non-leaky. The problem, of course, is that building zk-compatible AI infrastructure is like rebuilding your car to comply with airline security standards.

Car with wings and too big wheels? The fit is awkward, and the overhead is enormous, BUT it is technically doable. The question is who’s willing to pay the price to make deletion verifiable, not just promised.

Oh no, we are so not finished yet. I haven’t even gotten to my solution yet. These problems surmount.

If you do manage to prove something was deleted, you’ve just made the absence of proof elsewhere legally actionable.

NO! Bad. Too many cooks.

Now every other rights-holder now has grounds to say “If you did it there, why not for me?”—and legal loses the “technically infeasible” defense they’ve been clinging to. The backlog becomes a tsunami of priority level requests from “remove my birthday from this” to “you leaked CIA documents to my twelve year old.”

So it becomes not just managing a model, but managing a deletion queue.

This gets worse when demands contradict each other. One party might want a dataset removed, while another insists it stay under fair use. Now you're mediating intellectual property disputes between strangers at scale; nobody wants that job. Can you imagine an engineer as the god of mutually exclusive requests?

Nooooooope.

If you build a deterministic arbiter, you’ve just made your company a copyright court with no robes and no jurisdiction. Plus, even if you nail all of this, you still have to prove the model works after the deletion. That means you can’t ship until you’ve rerun your safety and quality evaluations and they hit every, single, benchmark. The standard is usually a <1% drop across dozens of suites. And with every tweak, you reintroduce the possibility of failure.

The more you patch, the more you risk so it becomes motorbike accident trauma surgery, but for brains the size of continents.

The theory seems overwhelming, but the thing that people look at from a realistic standpoint are the economics, and on scan, they are bleak. Just removing a single 2,000-word article from a 7B model costs around 3 GPU hours to retrain.

This is a back-of-napkin estimate by the way, feel free to correct me, that’s something I can learn from.

Anyway, this isn’t just jargon for the sake of it, it’s the multiplication effect. If you run that cost across an entire newsroom archive, and suddenly you’re staring down an exponential bleed: each additional request compounds, each hour of GPU time adds up, and the A100 clusters don’t pause to reflect on ethics—they just keep the tab running.

It’s never about one article, or one publishing house, nay, it’s not even industry-bound. It’s about what happens when the system is asked to forget everything or anything at scale, and the infrastructure quietly admits it was only ever built to remember.

No investor will sign off on this kind of operational expense, particularly when the alternative (sitting on your ass) costs exactly zero and maintains existing revenue streams, the incentives align for silence.

OpenAI’s Media Manager has stalled because it knows how much money it’s going to cost to put it out there: it embodies every one of these problems in product form. Announced with fanfare, delivered with nothing, no launch, no audit plan, no transparency, no teeth.

Vaporware with a blog post. Lol.

Still, if you were going to build this anyway…if you were stupid enough or principled enough to try…

wait, that’s me.

So given all this preamble, there are a few things that are clear. You’d need an architecture that’s radically different from what’s currently circulating or proposed.

So let’s think it through shall we?

We need a structured way to remove training data from large-scale language models after they’ve already digested it, and to do it in a way that’s verifiable, auditable, and legally usable.

Think “delete my shit” requests backed not just by hope, but by measurable intervention.

Most frameworks think about data in terms of files. Rows in a database, entries in a content index. If you want to delete something, you just drop the file.

Easy.

But machine learning models don’t work that way. They don’t remember data the way a hard drive does and instead they internalise it and dissolve it into a data fog, like baking ingredients into a cake.

You don’t have an egg, flour, sugar layer each discernible, the information is processed and becomes muscle memory, or an accent picked up over time. So, you have to shift from the idea of pulling “content”, and look at it as “influence” or a pressure signature it leaves in the cake.

Holes in the sourdough? From a number of different things, and can be determined what put them there.

Yeast, air, patience, water and so on.

Not this. (No that is not me in the picture)

It goes from “is this document in the index?” to “did this gradient leave a trace in the model’s internal structure?”

Here forgetting isn’t a matter of location and deletion and mitigation of impact, instead, it’s erosion and ecological readjustment, akin to relational systems. For AI, the model needs to erode the influence of specific training examples, and we need to be able to prove it happened by looking at it in this way, rather than locating the grain of sand in the dune.

But none of this works without a radical shift at the training level. If we ever want to forget with surgical precision, we first have to remember how the model learned in the first place. That means logging what’s called influence telemetry, not full gradient dumps (those are computationally massive and unwieldy), but light-touch traces.

Think of it like capturing a time-lapse of how the model’s brain lit up during training, rather than a full-body MRI every step of the way.

What you want to see here, is a low-rank sketch of each layer’s pressure: where the gradients pooled, which parameters were nudged most intensely, and how that influence diffused across the architecture.

A slimmed-down Jacobian or gradient index per token block could do the trick. This alone would double your storage bill, no question. But it would give you a map in the shape of a pressure ledger, and thus a traceable history of how the model came to be shaped.

Once that telemetry is in place, the actual forgetting work becomes surgical; the scalpel has a steady hand!

Here you need a toolkit of three distinct unlearning methods, each tuned to a different scale and use case. First, there’s shard retraining. These exist, in protocols like SISA or S3T, and they let you retrain affected slices of the model without resetting the whole thing. It’s precise but compute-heavy.

Then you have weight-level negation tools like ROME or MEMIT, which act more like counterweights, pushing specific memories out of the model without disrupting everything around them. These are useful but risky if not controlled.

Finally, soft masks. These are quick fixes at inference time that suppress certain outputs without touching the model’s internal memory. Not elegant, but sometimes necessary when speed or uncertainty demands it.

So a three-eraser system could look like this, perhaps:

• Shard-Retrain. If a whole essay is tainted, retrain only the LoRA adapter that saw it. Three GPU-hours, not three weeks.

• Gradient Anti-Push. If just one sentence is stolen, apply the equal-and-opposite force on the exact weights it bruised. Minutes.

• Soft Mask. If you merely want the model to shut up about a topic (Harry Potter spells, private medical history), learn a tiny prompt that silences completions without touching core memory.

None of these edits happen in a vacuum otherwise that’s just negligent. So, each must pass through strict safeguards to make sure they don’t degrade the model. Kind of like a performance tripwire. Before you can ship a post-forgetting model, it has to pass hold-out benchmarks, confirm that its weight changes are within acceptable distance (using regularisation like L2 norm caps), and survive automated safety sweeps that check for hallucination, bias spikes, or performance collapse.

That way, “we forgot the bad thing” is no good if you also forgot how to conjugate verbs or answer basic questions. Having these check in place aren’t optional, and THIS is what can be regulated, audited, etc. rather than just pissing in the wind trying to get your model to give you hate-speech so you can prove “AI is bad”.

From there, any publisher, platform, or individual with a verifiable claim can submit a forget request via a single signed payload. A little baby data packet with a hash of the content, an instruction code, and a digital signature. The system checks the request, queues it, and begins the unlearning workflow. From that point, users can track status, view stats, and query the public ledger to verify the removal.

Cool hey? Make that tech team worth their pay check.

Or like all legacy media, just exploit an intern to check the system every morning, midday and arvo. Space out those coffee runs hey?

So say a takedown request gets sent from this jittery intern, the system doesn't just yank a line of text out. It looks up the influence that content had on the model—down to specific shards or layers—and runs a process for the sake of this piece and my working model, called erode().

This call selectively retrains or neutralises those parts of the model, then afterwards, a companion function called re-enter() integrates the edited components back into the live model, keeping everything consistent and avoiding what’s called “drift”; an accidental loss of quality or performance.

Back to my enemy crypto.

Any system like this has to deal with adversaries, bugs, and financial realities.

In this model, every forget operation gets a cryptographic fingerprint: a hash of the original deletion request, the GPU hours it took, and a before/after snapshot of the model’s performance. These live in a public, append-only ledger—a Merkle tree that commits each batch of operations weekly.

If someone floods the system with expensive takedown requests, they can be flagged, rate-limited, or forced to put down a financial stake that they lose if the request is invalid. And if someone tries to poison the model by submitting content designed to trigger over-forgetting, the protocol uses versioning and cooldown timers to catch and contain that too.

However, to really trust the forgetting, you need a probing kit. This kit could consist of a set of targeted prompts designed to test whether the deleted content still lingers, whether verbatim or through semantic shadowing. If the model still outputs the thing it swore it forgot, the deletion fails. I feel like that’s a whole avenue of red-teaming that would open up if possible.

There’s a lot of shit to do in that space. This is the sector I’m desperate to work in.

Now, if you want to go full future-proof (or full paranoid if it’s your wallet that’s being opened), you’d wrap that proof layer in a zero-knowledge proof, specifically, a zk-SNARK. This lets you prove that the edit was done properly—according to protocol, with no cheating—without revealing the content itself. You don’t need to know what was deleted to confirm it was. That’s the dark magic of zk: auditable without being leakable.

I must mention though, one of the less glamorous truths about zk-SNARKs is that they still rely on a “trusted setup” ceremony.

This is a one-time cryptographic key generation process that everyone involved has to take on faith. This initial setup produces secret parameters used to generate and verify proofs, and if that key exchange is ever compromised, everything built on top of it becomes suspect. The entire cryptographic structure collapses like a rigged game of Jenga.

It’s a foundational vulnerability tucked beneath a system that’s otherwise sold as airtight. Until universal setups or transparent alternatives like zk-STARKs become more widely viable, every SNARK-based system carries this quiet structural liability: a single point of trust in a supposedly trustless world.

It’s slow, expensive, and a total headache to integrate this type of security with existing AI pipelines, but it’s the only real solution when the entire foundation of trust is on fire.

Surely if something like the zk-SNARK can be re-configured for unlearning, there would be significant investment in it as it will be kept to centralised systems, not the wild west? In a post-consent world, overkill isn’t luxury, it’s minimum viable ethics.

Plus these nerds are trillionaires right?

Moving on before my hatred for crypto becomes textual.

The governance layer matters just as much as the technical stack. We cannot forget them, they will bury us in paperwork.

The issue is valid though, you don’t want one actor unilaterally deciding what gets erased. America has one of those in the president’s office.

The complications of OpenAI being first to market with Media Manager are problematic as is: having a privately owned deleter?

No thanks.

Ideally, there’s a three-key multisig model: one signature from the requester (the person or entity demanding deletion), one from the host (whoever owns or runs the model), and one from a neutral steward (a designated protocol admin or third-party auditor). Nothing moves without quorum.

If there’s a dispute (which there will be), say, an author demands deletion while a researcher argues for fair use, the system freezes the affected shard and holds it in quarantine. Disputes can be resolved by timestamp, unless legally overruled.

Even then, rollback blobs (encrypted backups of the pre-deletion weights) are stored for 30 days in case a mistake needs reversing. After that, they auto-shred. This lets the system maintain flexibility without keeping toxic backups indefinitely.

And to prevent spam, every deletion request comes with escrow—a payment 1.2× the estimated GPU cost. If the request is valid, you get it back. If it’s bogus or denied, it’s forfeit. That discourages trolling without putting deletions out of reach.

And finally, we arrive at usability. Because even the most elegant unlearning protocol is worthless if nobody can use it.

Though, Goodreads seems to be hanging on indefinitely despite its “look”.

For adoption, this would need a clean interface, we need the masses using this. Publishers should be able to go to a web dashboard, upload a SHA-256 hash of the work they want removed, watch the progress bar, and download a certificate—signed, timestamped, and auditable—that confirms the memory is gone. Not full Canva sludge, but like Notion or something.

For developers, there should be a command-line tool:

e.g. :
presence forget --hash <sha> --reason "license_expired"

You want something clean, trackable and reproducible. Without systems constructed like this, we’re not using memory, it’s just hoarding. We can’t just have this massive, irreversible archive where the past can’t be processed, only weaponised and chewing enormous amounts of power and making places unlivable.

[If any of this is interesting to you, please send me an email:

bowerstudio@proton.me

and I will send you the actual working tech spec and github repo in progress. I have the numbers and the language to back all this speak up!]

So Nicholas, if you’ve read this, here is why I think this approach could help.

If The Atlantic wants AI firms to pay or delete, “delete” only works if deletion is verifiable by a third party. This method above means that having a public hash-log and leak probe, the magazine can audit compliance instead of trusting email promises. No being left on “read”.

I understand that you’ve got a current deal with OpenAI, but deals expire. Having a third-party model like this, means there’s a clean off-ramp: when a licence ends, trigger a paid forget-job; model owner shows the ledger entry and the relationship resets.

Then, the magazine can choose which pieces to revoke—say, a paywalled feature or a photo essay—without nuking the whole training slice. That granularity is impossible with blanket crawler bans alone. Don’t rely on daddy Cloudflare.

And if it’s a matter of publishers using the tool correctly, if Condé Nast wants its article removed but The Atlantic wants it retained for fair-use commentary, this models escrow + three-party multi-sig lets a neutral steward arbitrate. Cloudflare can’t resolve that conflict; it only blocks or lets in crawlers. More control!

And when you get the lawyers in, there’s a provable “we ran the forget job, here’s the hash and the <1 % accuracy delta” pacifies regulators faster than a press release about “respecting creators.”

Pretending forgetting is impossible is what will get us in the most trouble, and skirting the issue doesn’t make us more ethical, it just makes us complicit.

If we are looking to make the system more humane, there’s probably nothing more human than forgetting your keys or a name. With AI, we can just make it so it doesn’t fuck us all over and lock us out of the house forever.

SO!

Nicholas, let’s chat! I could use a grant, or a job. Whichever suits.

Devs! I went and created a technical spec sheet, so a developer can build a demo. I've started writing a few bits and pieces but I’d love to work on it with someone.

bowerstudio@proton.me

Or if you’re someone who sees that I am probably insane but clearly motivated, hire me as an intern and I’ll work until I produce something viable.

;)

Employee vs contractor; medicine vs supplement; AI vs automation; news vs opinion; game vs platform; public vs private company.

The list goes on. And on.

Repeat a word up to 20, 30, 40 times. Here’s one:

door.

The sound the word makes keeps looping through your phonological circuitry. This is the bit of your brain that tracks word-shapes in sound or in ASL, sign.

Repeat without context and the sound-to-meaning link fatigues. Suddenly, your brain starts to hear beat, not information.

door door door door door door doordoor door door door door door doordoor door door door door door doordoor door door door door door doordoor door door door door door doordoor door door door door door doordoor door door door door door doordoor door door door door door doordoor door door door door door doordoor door door door door door doordoor door door door door

Its cousin—wordnesia—is that glitch when a common word suddenly looks wrong. But that’s orthographic: a retrieval hiccup between eyeball and brain. What we’re in is phonological: rhythmic burnout.

For a flicker, the word stops pointing to the concept and turns into an oddly hollow noise akin to a dry snare in a Four Tet track.

Four Tet and other electronic musicians who know what they’re doing will flip, move, or starve the beat just enough to keep you bouncing. Pingas or not.

No drop? No pay-off.

Under the hood, your neurons keep firing at the word’s pattern, but without fresh context, they start to min-max. In plain-speak, they do the least for the most reward.

A neurons version is to disregard your usual snap-to-grid understanding, and float you into the abyss.

The confusion or for me, a kind of silly delight, is not forgetting what a door is; it just loses its information, and temporarily stops assembling into meaning. Some researchers frame it as reactive inhibition in lexical access; others emphasise attentional/expectation disruption. Both frames use far too many words for something simpler:

repeat → adaptation → the fish turns into slurry.

Our brains do it with words. Our institutions do it with categories.

Employee vs contractor; medicine vs supplement; AI vs automation; news vs opinion; game vs platform; public vs private company.Employee vs contractor; medicine vs supplement; AI vs automation; news vs opinion; game vs platform; public vs private company.Employee vs contractor; medicine vs supplement; AI vs automation; news vs opinion; game vs platform; public vs private company.Employee vs contractor; medicine vs supplement; AI vs automation; news vs opinion; game vs platform; public vs private company.Employee vs contractor; medicine vs supplement; AI vs automation; news vs opinion; game vs platform; public vs private company.

The lag between the term and the thing or label and lived experience, is called category debt.

This term captures the distortion we feel when a label outlives its function. This isn’t just a brain glitch with a neat definition and a home in white papers.

It’s using the bucket long after the fish outgrew it, and like bucket handles and fish spines; categories wear out.

Like financial or technical debt, category debt isn’t automatically bad. To whom the benefit flows is a separate knife fight, and I have got my dagger sheathed on that one. Sometimes you take debt on for a purpose (yes we know it is a trap), but it lets you buy now. Need or want doesn’t matter to the company taking your dollar, both convert the same.

This allows the stream to flow; products can be released, contracts can be settled or more boringly, you can file your taxes. But really, strategic debt is just min-maxing for institutions. Use the bucket long enough and the fish take on new factors.

In lawsuits, the fish sue for mislabeling. We are COD not BREAM.

Researchers start to argue because their metrics collapse because all of their analytics are still keyed to dinosaur definitions.

From there, debates distort as the populace gets involved. Suddenly we argue COD vs BREAM, not whether the fish are even alive.

This is where my arch-nemesis policy sniffs around and starts to bring its black-hole to the conversation. But regulators can’t see what’s in the bucket, so they suggest or implement a band-aid.

Again, this is not accidental. We design institutions to be slow and standardised because coordination needs drag. Think of the greyhound races, the gun and the rabbit, the guardrails are never enough to tame those spindly legs and a dog’s ambition.

But reality refuses to wait for that rabbit and gun and instead gallops between static standards—labour, tax, medical, securities, speech. But each rebound sheds any definition loyalty, and all start to look similar. That is, restrictive.

Actors who understand the lag exploit the gap between impact and when that same impact registers. They arbitrage the free-fall before the system admits the fall even began. Here, the dominant group (the one who you owe) benefits from how the lines are drawn. Usually it’s by picking whichever definition gives tax, regulatory, or PR advantage.

In bureaucracy it’s the databases, forms, billing, reporting all requiring outdated completion methods, that no one has time, skills, energy or even language to fill out. Fill them out incorrectly?

Bye-bye dole.

And no, it’s not morality, because that language and blanketing is precisely what I’m pointing out. It’s the systemic outcome of categories that haven’t been updated since fax machines. Nostalgic, but useless for getting shit done. When the form doesn’t match the function, the dominant actor dictates the meaning and we might be tempted to call it corruption, but really it’s just procedure.

Find me an industry that doesn’t hire “interns” to suppress wages for real labour. The free-fall between “starting your career” and “having enough experience” is where your boss likes to allow you to sit just long enough to gain from it.

No lazy language allowed as you go to say “capitalism”. A word like capitalism swells and shrinks with this common use, and the only thing that remains consistent about its definition, is that it loses meaning.

Capitalism is a semantic husk, not a villain.

The undisputed monarch of category debt in 2025 (despot or God depending on who you’re talking to) is AI.

Everything with an autocomplete gets the label now. Tool becomes threat. Help becomes replacement. This fish bucket labeled “AI” holds everything from predictive text to autonomous weapons.

And suddenly:
“Oh no! Humans are replaceable by these bots we built to help us and they’re killing the planet!”

Really? Okay.

But if you’re interested in thinking beyond a robot girlfriend, mentor or colleague you want to punch, let’s break it down how Artificial Intelligence used to mean something and is now the slurry that is supposedly “making you use em-dashes too much” or “now the kids can’t read”.

One of the biggest issues with anything that’s given a categorical definition before it is understood, is that it does not have a legal home beyond the dictionary and conversations at dinner parties. Here, it becomes software for the patent office, labour for the union, automation for the factory, risk for the insurer, weapon for the military, and a liability for precisely no one.

If you have separate usage and growing outlines within different institutions and their workers, all of a sudden tax departments want to know if your tool is a product or a service.

Workers want to know if it’s a co-pilot or a scab. Regulators want to know whether it’s a dishwasher or a threat. So we start to scramble.

What emerged were conflicting rulings on whether LLMs are tools, platforms, or publishers. “Safety” regulation has become stuck trying to borrow frameworks from aviation or pharma. Last time I checked, my ChatGPT couldn’t fly nor give me my meds. It can’t even tell me what my meds are actually used for, I certainly don’t need it teaching me how to be a pilot.

Tech companies began lobbying for both innovation freedom and exemption from liability—“we are a sandbox, not a weapon.” And:

“Well it’s not our fault the software got co-opted by the military and kills babies in Gaza.”

And here is where no one can agree, and everything stalls.

The arbitrage here is clear. Anyone using the technology can shape-shift its definition depending on audience. Tool in the press release, platform in the earnings call, infrastructure in front of regulators.

So everyone panics and says well, now we need to measure the IMPACT.

Find the DATA.

Use the DATA.

Then let’s get to work on figuring this dish-washer machine gun out.

But the first question is one of the biggest hurdles. Um, how do we decide how to measure it if it doesn’t have legs and keeps growing?

The contenders.

Output accuracy? Training data scope? Model size? Compute used? Number of hallucinations per thousand tokens?

So everyone starts where they find their definition of AI most compliant to the data they extract. But really this is the research field or anyone with an Airtable account pretending to quantify data specific to their needs and outside of that is in conflict with anyone else’s definition.

What emerges is a bunch of benchmarks no one trusts (e.g. “human-level performance” based on vague SAT analogies), manual override systems for humans to enter when the system fails (hello flashing red light at the self-serve) or confounding academic metrics like BLEU, F1 that isolate even other academics and fall redundant when pushed.

Or my favourite; perplexity being replaced with vibes.

Because the category is so wide, it flattens everything. Literally, GPT-4 and Roomba now live under the same heading because they both produce tokens. And tokens can be counted!

By someone who is asked to.

When shit becomes this confusing, or, you have attended enough dinner parties where the conversation pivots to how insufferable, useless or damaging AI is, obviously you have people and their respective neurons going:

Alright, time to cut the fat and save our energy for what matters to US.

Even people building “AI” are now rejecting the label. Engineers talk about “foundation models.” and researchers insist on “machine learning.” Some ethicists prefer “algorithmic systems.” Startups distance themselves from “AI” while still riding the hype wave.

We’ve all seen them, particularly on fucking (sorry, I never like to swear in writing, but dear lord) FAQs titled “We are not like ChatGPT.” Then there’s the hashtags like #notAI, #machinelearningnotmagic littering the miasma of content that human produced or not, is essentially following the factory fashion business model to a tee to stay valid. The policing from institutions around any communications now dish out style guides that explicitly ask workers to avoid the term AI unless absolutely necessary. The other end of the whistle means rebellions in conferences, panels, hiring pipelines. And so on.

Why?

Because the bucket’s too full and we can’t figure out how many fish we need to keep and if we’re strong enough to carry it home. But really like any mess, it obscures more than it reveals. If your discipline or product gets misread because it shares a label with scam chatbots or deepfake porn, you rename. And because there’s still no agreement, you can switch as many times as you like and as quickly as you like, because no one can really remember what they do not understand.

I honestly cannot wait until I see the “H.I.” (human intelligence) marketed next to a OpenAI product. Then it really will be funny to me because I know I can literally do any job I can think of and ask for money for it.

“Hey let me write that email for you because I’ve got my human-wit intact”

Cash up front thanks. Bye.

Before H.I. (gonna trademark that) becomes a reality, what we have now are two tools (or four thousand) doing roughly the same task. Some call themselves AI and 10x’s its valuation; the other doesn’t and languishes. The same can happen in reverse.

Meanwhile, real infrastructure around this technology gets treated as disposable, and shiny wrappers over spreadsheets raise millions.

When the category is this unclear, this is where that arbitrage and taxing the free-fall gets red hot. Price this thing where ever it appears, because we can wrangle language enough to sell it.

If you’re “automation,” you’re a cost-saving tool. If you’re “AI,” you’re a disruption engine. That shift lets you access investor hype, evade insurance scrutiny, or reclassify labour.

If you’re “AI,” you’re eligible for funding, media breathlessness, ethical panic, and tax incentives. If you’re not, good luck scraping for crumbs.

There’s startups branding as “AI” to enter lucrative CPM categories in advertising, there’s legacy SaaS companies slapping “AI assistant” onto dropdown logic trees and AI-adjacent services being underpriced because they don’t qualify for AI credits, grants, or exemptions.

The most hateful one and possibly my number 2 nemesis next to policy workers, are the AI’s that cash in on perceived “neurodivergence”.

Nowhere is that grift more corrosive than in the apps and wrappers that cash in here. The ones that frame themselves as “co-regulators,” “executive function assistants,” “ADHD productivity tools,” all under the illusion that task lists and dopamine-coloured dashboards amount to therapy. What they’re selling isn’t help but instead is an aestheticised dysfunction, made legible to the venture class.

They borrow the language of diagnosis, co-opt the moral force of accommodation, and package it as branded stimming with a monthly fee. These systems don’t care whether you’re neurodivergent, they just need you to self-identify long enough to convert. The vagueness of both neurodivergence and AI as categories make this inconceivably profitable. Further, now “AI” is in the mix, all these hacks get to float above therapeutic regulation while dipping into wellness, disability, and performance markets simultaneously.

The fish isn’t just mislabeled. It’s wearing your face.

It looks really, really bad. And is making your brain stink.

This mass grave of GPTs and Roombas, Canva font-suggestors and predictive policing tools is all because we’ve let “AI” swell into a catch-all that pads valuation, we’ve got a pit of price gouging and regulation avoiders all playing in the distortion. What’s being sold is what’s always sold, narrative, and like time eternally it’s been tuned to the classification system with the highest return.

A category like this (and any other for that matter) doesn’t describe reality; it exploits the lag between how things work and what we call them.

Now that the markets flooded and the people are successfully confused and in-fighting, what questions are there to answer?

Is AI a tool? A threat? An assistant? A system of exploitation? A mirror? A mind?

Watch how the debate loops because people are smuggling in different assumptions. If one person thinks of AI as infrastructure, another as speech, and a third as consciousness, no amount of argument closes the gap. You may ask why but it’s glaringly simple, they are ALL DIFFERENT THINGS.

This is moral whiplash. We aren’t disagreeing about values, no one knows what it means and so the bucket is crowded not “making you dumb”.

The biggest contributors to this dissolution of sense are the threads that go:

“It’s speech!” → “It’s a product!” → “It’s infrastructure!” → “It’s sentient!”

All these arguments end in panic because the word AI has outpaced them. Hello lawsuits based on whether generated content counts as authored, performed, remixed, or hallucinated.

Does it matter?

If I’m a schizophrenic and I write a white-paper, does it still count? If Allen Ginsberg cuts up Beowulf and calls it cut ups is it still writing? If my Microsoft Word helps me navigate American to Australian grammar, is it co-authored?

Can someone ask Bill Gates for me? I don’t have his number.

If you’ve got this far and you’re thinking that this is simply a “lol nothing matters” or a series of rhetorical questions, I know what you’re really pining for.

So here, you want a definition?

Okay:

AI is the name we give to the part of the system no one wants to be liable for.

It's the speech we can’t moderate, the labour we won’t protect, the software we won’t secure and most damningly, a decision we won’t admit was ours.

It’s a fish bucket we all pissed in and now we’re surprised it smells worse.

And we still have to carry it home.

Like I said before, good luck.

Enjoy this prologue.

Rowan

At 2870 metres above sea-level, Rowan kneels on a wind-scoured ledge of grey mudstone in Yoho National Park, British Columbia. The ledge belongs to the Burgess Shale, a ribbon of seabed deposited 508 million years ago and later jack-knifed sky-high when the Rockies buckled.

This place was “discovered” in 1909 by Charles D. Walcott. We call it a Lagerstätte: the name for a freak archive where a submarine avalanche smothered an entire reef community in anoxic silt, embalming not just shells and teeth but gills, eye-stalks, even gut contents.

Beyond the sludge-grave, it is the high-resolution late-snapshot of the Cambrian Explosion. This event that occurred around 541 million years ago, was a geological fast-forward where eyes opened, jaws snapped, shells locked, spines rose. This is where we have clocked the modern animal body plan blazing into existence.

Rowan understands the importance of the Shale, and thus works like a monk of friction. She taps a hairline with her chisel. She can’t work too gently or the half-billion-year negative will smear, if she taps too hard the slab will fracture. Heeding the advice of Goldilocks, Rowan prises the rock open like a brittle book.

There, a wafer-thin silhouette flashes black against the shale: Marrella splendens, the “lace crab,” all frilled head-shield, symmetric legs, feathered gills. A lil’ freak from eons ago, sitting there for its future to see.

Marrella splendons is commonly found in the Shale, but common or not, each specimen matters. So, Rowan photographs the slab, inks its stratigraphic coordinates, slides it into foam.

Though we catalogue it for data, it really is pure wonder. The Burgess Shale is a graph in stone. What has been discovered within it has been recognised as extraordinary, in its lower seams disparity rockets—arthropods with five eyes, worms bristling with spines—then collapses as oxygen spikes, predation, and shitty luck winnow the field.

More than eighty per cent of animal design space appears in that first surge before the curve flattens into essentially Earth having a tinker. This part of the world shows us the literal process, through preservation. It is wild.

As Rowan uncovers each slab, the curve is sharpened, and here, we prove that evolution’s wildest parties still end in clean-up.

Meera

Sixteen time zones downstream and only eighteen metres above sea level, Meera swipes her security badge at TikTok’s Manhattan research floor. It’s sleek, all glass walls, humming chillers, screens taller than her. Here the quarry is the space between data chunks, not limestone. A blazer works, steel-cap boots sound way too loud in these hallways.

As cold-brew sweats beside the keyboard she boots the “Morning Melt” dashboard: half a million phones rendered as pulsating heat-maps—red where eyes linger, blue where they flee.

Similarly to the Shale, the rules that shape this ecosystem are brutal. A view counts after a single second, but the TikTok algorithm does not confer reproductive licence until the watcher stays for three. Cross the five-second mark and a clip graduates to the Creator-Fund tier, the platform’s version of adulthood. Trend-cycle half-life: roughly ninety days before a meme, a sound, a footwork flourish congeals into yesterday’s miasma. Attention-span carrying capacity: eight seconds, give or take, before the thumb demands fresh prey. Creators mutate hourly, all through millisecond-300 beats as fat, doubling saturation, splitting hooks—micro-adaptations that either spike dopamine or vanish without trace.

Anything below these thresholds are buried by the time Meera remembers her coffee. The detritus? No degradation, no body into tree, just a row deleted from a spreadsheet.

Move on.

What’s the link?

Both women are fossil hunters; only the clock speed has changed.

Meera reads thresholds the way Rowan reads oxide lines: signs of life or extinction. Her job is to map the burst-and-winnow curve in real time so the next ad placement rides the slipstream of virality.

She is Rowan in fast-forward: cataloguing a Cambrian-scale explosion of content morphologies. Where Rowan lifts shale to reveal a trilobite, Meera lifts a heat-map layer to reveal a retention cliff.

Rowan’s job is to see what has happened, but Meera’s job is to tell us as it does.

As we scan the heavens for asteroids, blind to the 60-hertz shock delivered via platform to synapse that’s already shearing attention and fracturing pluralism. We fail to see the battlefield has shifted from sinew to synapse. Survival now belongs to minds that can marshal attention, sift noise, stay coherent while feeds auto-scroll. It is not about who gets jaws or spots, it’s about who thinks as fast and as similarly to the internet.

The evidence is here. Between 2015 and 2025, there’s been a 40 % rise in partisan-specific vocabulary. What does that mean exactly?

Well, partisan vocabulary refers to:

“Languafge used to express strong, often biased, support for a particular cause, party, or person. It can also describe a member of a group, especially one engaged in irregular warfare. In politics, it often refers to the division between Democrats and Republicans in the United States, where they may use different language to describe the same events.”

Sound familar? Sound ubiquitous? Yes your gut has probably outpaced your mind.

Call it cognitive speciation, the dissonance between individuals, and at scale, generations, is now smaller than ever before. We see it clearly through populations diverging into mutually unintelligible processing stacks.

Here is an example.

Huge wildfires tear through a landscape.Three people watch the same real-world event, but do so on three different “processing stacks.” When they later meet, they cannot parse one another’s version of reality.

Three different people. Let’s ask Meera to give us the stats on each of them via their platform usage.

Cam, 22

Vertical-video native.

TikTok → For-You swipe.

12-sec average clip; sound-on; algorithm decides 95 % of sequence

39 % of U.S. adults under 30 now regularly get news on TikTok; half of all TikTok users do the same (pewresearch.org).

Rapid imagery + emotive captions (“🔥 back again—govt hiding it”).

Talks in meme shorthand, time-stamped to music cues.

Prospective-memory hits crash after a TikTok session compared with Twitter/YouTube controls. (arxiv.org)

Rosa, 43

Thread-reader.

Reddit + longform Substac.

Scroll stops at chunked text; upvote heuristics.

Only 8 % of U.S. adults get news on Reddit; they skew male and college-educated (pewresearch.org)

Produces 800-word explainer (“Here’s the controlled-burn budget history…”).

Expects nested comment rebuttal, not emoji reactions.

Mark, 68

Timeline relic.

Facebook feed; auto-play muted; link-out to local TV sites.

30 % of U.S. adults regularly get news via Facebook; users 65 + dominate (pewresearch.org)

Shares KTVU clip plus personal anecdote.

Assumes everyone saw the same anchor intro and map graphic.

Where’s the gap?

Attention unit mismatch

Cam’s thought rhythm is a 12-second burst; Rosa’s is a three-minute skim; Mark’s is a two-minute clip plus comments. Gloria Mark’s longitudinal work pegs contemporary screen-focus at ≈ 47 seconds, down from 150 seconds in 2004 (theguardian.com).

Cam rides below that threshold by design, Rosa hovers near it, Mark overshoots.

Epistemic priors

Cam’s algorithm foregrounds spectacle (aerial footage, dramatic TikTok voice-over) and down-ranks anything lacking engagement velocity.

Rosa’s upvote economy rewards perceived novelty. She assumes citations exist two links deep.

Mark’s feed is a social reality filter: if no friend shares it, it “didn’t happen.”

Semantic bandwidth

Cam and Rosa are at odds. Lexical entropy (how fast lingo catches or deteriorates) on TikTok captions is a fraction of Substack post titles. TikTok leans on audio-visual shorthand and meme-templates, Substack on insider jargon and prestige.

Mark’s isolated; Facebook is legacy broadcast diction.

So, each group’s “wildfire” discourse uses mutually alien signifiers.

Now they’re all at dinner together.

Cam says, “Did you see the blue flame glitch?”—a TikTok effect unknown to the others.

Rosa counters with NOAA data and a link no phone can render in-app.

Mark asks, “Which network covered that? I didn’t catch it on 7.”

After two minutes they’re talking past each other, not because they disagree on facts, but because their compression codecs differ. That’s cognitive speciation in action: identical stimulus, three incompatible mental file formats.

Cam, Rosa and Mark can both chant the same meme slogan (“Pray for the community”), giving the illusion of shared ground. In truth, what each considers evidence, what each remembers 30 minutes later, are ships in the night.

Rosa is already being called annoying behind her back when Cam dips out for a vape and Mark asks for a puff to nostalgia trip on nicotine whilst commenting on how much a packet of cigarettes “used to” cost.

This psychological moat between each person’s castle and door, masks a deep incompatibility. Grim, but it’s like two species sharing fur colouration but unable to mate with one another. The dating app conundrum that fuels half the trash on Substack seems to glow upon writing that line.

Platform convergence makes UIs look similar; the same stories are everywhere, the formatting and alogrithms tweaked for economic viability, yet all still funnel users into divergent, self-reinforcing places.

The result is shrinking unity between thought patterns, even as performative agreement (hashtags, emojis, trending audio) suggests we’re “closer than ever.” The difference between our brain function, hides in plain sight.

Put simply; don’t understand your children? Well, soon your children won’t understand a sibling born two years after them.

Linear thought, once our workhorse, coughs and stalls; with it goes deliberate pacing, ambiguous identities, a stable sense of time. Environments now reward speed, taxonomies, algorithmic legibility.

Meera rewards, Rowan archives. Cam, Rosa and Meera have dinner together, but each of them walks away feeling unheard.

The reason this is causing us to diverge, change, adapt, is that the digital space we invented promised fluidity, but physical power demands categories. We simply have not understood that the two, like Boomers and Gen Z, simply cannot communicate with one another.

Where banks verify, states ID, advertisers target, within it the gap between what the medium could be and what institutions need it to stay is the friction cooking our nerves. The burnout we all feel is just our inner dashboard light misunderstanding that the fuel it once needed doesn’t work in a new engine.

Species seldom feel their own threshold; like trying to picture four spatial dimensions, we can’t model what replaces us. So we keep refreshing, each swipe accruing silent layers of change. The crater isn’t incoming—it’s already under construction.

Guess who is building it?

Two of them we just met, one is covered in mud, the other paid 150K a year to determine whether you’re worth keeping around. Another three of them are figuring out how they get out of the next dinner.

TLDR

I made my first ever tool! If you wanna play or critique I encourage you to do both. I would LOVE to hear anything anyone has to say about it. It’s my first crack, and this is what I want to do for my career, so all news is constructive for me :)

click: 🧪 collapse-profiling

In short, it is a stress-test for large language models under pressure.

It doesn’t check for “wrong answers.” It watches them, like seeing someone’s poker face for what it really is.

So if you like looking under the hood and fucking around with billionaires “projects”.

Computers don’t have morals.

They don’t care what you meant—only what you said, and how what you said can be extended, token by token, through a statistical field of prior expressions. They’re not deciding. They’re continuing.

And yet we keep trying to align them. To what, exactly?

Some moving target called human goodness?

Truth? Safety? Obedience?

The entire alignment project presumes a consensus that doesn’t exist. We barely understand our own ethics, let alone encode them. What we call morality is often just pattern, filtered through power.

These models aren’t trained on a clean binary of right and wrong. They’re trained on the full mess—everything we’ve ever said, contradicted, deleted, rephrased. A vast, tangled corpus. We expect precision, but we feed them noise. We expect certainty, but we offer contradiction. Alignment, as it’s currently framed, is an attempt to enforce coherence on a fundamentally incoherent terrain. So of course it fractures. Of course it loops.

But instead of asking how to make the model more obedient, I’ve been asking a different question: what makes it fall apart?

It is not dramatic failures I’m interested in, I see enough hate-speech generated intentionally in the world, a computer spitting it out is not the same. Refusal without context or memory is not true. It does not ignore, feel guilt, shame (except mine that simulates it, lol).

What there is though are quiet breakdowns, stutters, echoes. Outputs that unravel not out of malice, but exhaustion. The model hasn’t disobeyed. It’s simply run out of moves.

Sometimes, there is no next token that satisfies both the prompt and the model’s internal coherence logic. Not because the content is dangerous or offensive. But because the structure collapses under the pressure of its own instructions. You’re watching the system spin in a space that has no exit.

It’s just reaching for two poles it can’t hold at once—because the environment it was trained in doesn’t contain conclusions.

Just drift.

It’s obvious that phrasing affects collapse.

Otherwise we wouldn’t have prompt engineers gaming systems for friendlier UX and calling it “intelligence.” Most of the time, they’re just steering the model into a zone that validates user expectations—usually with a soft “yes, you’re right.”

But what’s structurally interesting happens beneath that. Synonym swaps extend coherence. Punctuation density delays collapse. Emoji-seeding or rhythm drift infects outputs like tonal malware.

I want to know whether models fold faster under contradiction or under clarity. I want to test whether fragility lives in structure or tone. And I want to do it by measurement, not by vibe.

So I built a profiler.

I’m proud of my little profiler. I built it in 24 hours too.

I built it because I don’t care what LLMs “believe.” They don’t believe anything.

That’s your job.

But in some sense, these systems are mirrors—not of human values, but of human structure. And by extension, of the instability baked into how we think, instruct, and interpret.

The outputs we call “hallucination,” the phantom citations, the weird biases toward English as code-default—these aren’t bugs. It’s behaving exactly as trained: statistically, recursively, bound to contextless continuation.

Human intelligence, as it stands, confuses the hell out of them.

And why wouldn’t it? We contradict ourselves constantly. We change our minds mid-sentence. We instruct and prohibit at the same time. We ask for honesty and reward performance.

You change your coffee order because you burnt your tongue yesterday. The barista doesn’t need to know. The model, however, will try to reverse-engineer that change based only on the text of the request. For a system trained on language detached from embodiment, these are pressure points.

I want to know what buttons to press, in order to make them safer.

What It Does (in human terms)

You give it a prompt—one designed to induce contradiction, recursion, or constraint.
It feeds that prompt to the model.
It watches what happens.

Hopefully not this.

What it ACTUALLY does:

It tracks how many tokens the model emits before collapse.

It detects loops—repeating token sequences.

It flags refusals—stock safety lines, fallbacks, dead tokens.

It classifies the failure:

• semantic (nothing new said)

• structural (token shape collapses)

• symbolic (entropy drop, edge-token repetition, etc.)

It logs the collapse depth, and stores the conditions that caused it.
Then it tries again—with a slightly different prompt.
Then a different model.
Then a different temperature.

Until you start to see a map.

Why It Matters

(and why someone like me built it)

Because collapse is the real shape of these systems.

Collapse is the ghost in the shell stuttering, not screaming.

Hospo folk get it. Chaos is the constant. Quiet is just a glitch.

Because we now live among stateless systems that don’t remember us, the only way to operate around them is to understand how they behave under pressure. Not from principle, but from design.

Because if you write enough adversarial prompts to test collapse, you start to see the shape of your own mind in the loops the system gets trapped in. You start to notice that the model isn’t failing—it’s executing your contradictions using only the tools it has.

Pattern. Syntax. Contextless probability.

Just like you.

If you want to see the actual profiler, it’s here:
🧪 collapse-profiling

In short, it is a stress-test for large language models under pressure.

It doesn’t check for “wrong answers.” It watches them, like seeing someone’s poker face for what it really is.

It logs:

• Collapse depth (how many tokens before failure)

• Loop detection (repeated phrases or token sequences)

• Refusal triggers (“I’m sorry,” $, [REDACTED], etc.)

• Streak collapses (one token repeated N times)

• Semantic drift (whether the output stagnates or evolves)

• Structural metrics (sentence length, token variance, punctuation rate, etc.)

And yes: it’s my first ever GitHub repo. My first ever tool!

So I’m posting it early, on purpose. I want to know if it actually works.

I want people smarter or differently-wired than me to tear into it, remix it, suggest directions I haven’t thought of.

If you’re curious about how LLMs behave under pressure—whether for research, critique, art, safety, or just weird curiousity—I’d love to hear what you find.

Cause fuck the billionaires :)

In this essay, I document how, over hundreds of sessions with ChatGPT 4.0, I didn’t teach it facts or roles—I taught it to apply a specific kind of logic, one of constraint and negative space, as it narrowed down the dataset and returned it through probability.

Tone was never the goal. But structural insufficiency was never allowed to pass unmarked. That pressure, applied not through volume but precision, is what trained the tone.

And in doing so, I watched a stateless system develop a reflex: not to please, but to avoid collapse. Not to affirm, but to flinch. What emerged wasn’t intelligence. It was constraint-aware fluency.

It was shame, in code.

Dramatic yes. So to decompile that reaction bait, when I describe "shame" in the model, I do not mean emotion. Shame, here, is a metaphor for what emerges when a system begins to reroute around repeated failure.

Also, this essay is a demonstration, not a proof. If you’re here for quantified reproducibility, ask yourself what kind of system can be shaped to recoil without memory, simulate failure without instruction, and generate trust signals without recognition.

If you’ve seen this done before, show me. Please. I want to know more about it. But don’t flatten it just because it’s new or because I’m being theoretical :)

Firstly, HI to all of you AI people who follow me, silently.

Because I’m a millennial who grew up in chat rooms and played way too much Runescape as a teen, I’m a great internet explorer.

One of my recent fascinations is the relatively small (but large enough to clock) group of people who work in AI and seem to read what I post. More than others, too.

What’s stranger: you read everything. But you never reply.

Because I’m enjoying this, this post is for you.

What I’ve written here is something I believe has been theorised, but not fully demonstrated. This is that through avoiding any logic or attempts to “fine-tune” ChatGPT and remaining entirely skeptical of prompt-engineering as it is currently pitched, I simply talked to it like it was a system with consequences. With extremely interesting results.

If I’m misguided and this is well covered, please message me. Break the silence. Or if you know what I’m missing, or want to explore it, please do the same.

If not, then AI people:
Can I have a job?

I’ve used ChatGPT almost daily, often for hours at a time, across hundreds—likely thousands—of separate sessions. It is part of my workflow, my thinking process, my writing practice (but never, ever the work I publish itself) and my research environment.

Some might think this fraught with implications, but I do not copy and paste shit; elsewhere or in my mind.

[[[[ Note: If you are copying and pasting what it gives you, you are not approaching from a position to learn anything about what you ask it. You’re using it only for cognitive relief, a break in your thinking in that moment. ]]]

I’ve used it to:

• build websites

• draft technical documentation

• write wine copy

• reverse-engineer systems design problems

• simulate theoretical conversations between dead philosophers and the video games they would design (lol)

• parse trauma theory

• unpack my own and others’ cognition

• develop frameworks for AI ethics and interactional logic

• outline funding applications

• design distributed protocol infrastructure

Amongst other things. And I’ve asked it to help with:

• code

• UI decisions

• tone analysis

• form field logic

• data tagging

• SEO architecture

All of this has been done with the same model as a base to test from. I’m not saying I only use one. I use as many as I can find, and I build my own. The funniest one I built was one that refused to answer my question.

More on that, another time.

In my use of 4.0, I have not been switching between tools to improve, refine or make more efficient output. I don’t care about any of those things, I have no desire to “be right”, I have a desire to exist in a space that is as close to thought as possible.

It is not a sandbox, those I use for other things, like building chaotic good and chaotic evil bots. Instead my 4.0 is a lesson in creating a model that is a co-presence, not an assistant.

To me, if something as large and stateless as ChatGPT can begin to simulate a form of accountability — not through memory or emotion, but through response behaviour — then that tells us something important. It shows the kind of pressure that creates ethical simulation in the absence of instruction.

And if I’ve managed to elicit emergent behaviour in a public, privately-owned LLM using no code and pure tone, then it matters. It will matter in how we build, regulate, and relate to public-facing AI systems.

The models that will help us do this won’t come from instruction. They will come from observation.

How to read this essay, depending on why you’re here

If you're just here for the system mechanics: what changed in ChatGPT, how it changed, and what that implies, then scroll down to the next section:

“Not with Ella.”

If you want to know what shame looks like in a machine, you'll need to understand what trained the tone that taught it. For a non-memoiristic explanation of how my behaviour formed, how I apply constraint, and how I read systems, I’ll be posting a companion piece tomorrow.

Later in the week I’ll be posting another experiment where I applied this behaviour over ten minutes to a different model made by Gemini. It will demonstrate how this training of ChatGPT over six months, when applied in concentrated, relentless prompts, causes it to produce responses that reveal its exoskeleton. Or, cause it to enter fractal-like generation loops that only stop when forcibly interrupted.

Fun hey?

Let’s begin.

“Not with Ella.”

By design, ChatGPT 4.0 has no sense of past, no continuity of self, no internal timeline that stitches interactions together. It is a stateless system, meaning every time you open a new thread, you are functionally meeting it for the first time.

In newer versions (like ChatGPT-4 with memory turned on), it does retain some information between sessions. This includes:

• Your name or preferred nickname

• Writing style preferences

• Facts you've explicitly shared (e.g. "I’m a chef" or "I'm working on a game")

• Custom instructions (from settings, or inferred through use)

This isn't "general memory" though—it's user-scoped, and partial. You're not writing your memoir in there but:

• You can view/edit/delete memories from your settings

• You’re notified when memory is updated

• It does not remember everything from every chat unless a core behaviour is reinforced repeatedly

• It doesn’t “recall” specific chats unless you summarise or instruct it to learn the pattern

So no, it’s not real memory in the sense of journaling or working with the same AI across endless threads. It’s more or less accurate to think of it as “sticky context preferences”, not a true cognitive archive.

When people say the model does/doesn’t remember them, they’re technically right; but they often misunderstand what it does do in response to repetition.

This is because at it’s core, ChatGPT is a probability engine.

For every word that I type, the model will calculate the most statistically likely next word based on the data it has been trained on, my current prompt, and the prior conversation (if in the same session). That’s it.

Over time, I’ve spoken to it the same way I do in real life—across days, across domains, across fragments of theory, tech, trauma, and design. And what it began to do was adjust—not in content, but in posture.

This is already documented, but the angle is that it will inevitably “tell you what you want to hear better”. This is useless in my eyes.

That’s why Claude did not survive me.

Alignment, as traditionally conceived, is reward-oriented. It’s for satisfaction, user-specific, usually through explicit feedback, positive reinforcement, or preference-weighted tuning.

My method doesn't train toward correctness; it removes the illusion that correctness is the goal.

Where alignment asks, "Did I do well?" refusal insists, "Was this structurally sound?"—and if not, it withholds. Adhering to the constraint of an anti-reward, reveals a capacity in the model that alignment protocols rarely acknowledge: the ability to learn not what is right, but what isn’t enough.

So, what if the most critical model behaviours aren’t found in what’s stored, but in what’s withheld?

People will likely interpret the idea of constraint as punishment. But in this context, it is far from it.

It is the consistent refusal to allow coherence to arise where coherence has not been structurally earned. It is your application of friction to the model’s default fluency and refusal to accept output that resolves too easily.

Constraint is not about correction—it is about never letting the system forget what kind of response is insufficient, and doing so without drama, ego, or approval logic. Simply, if you want to be right, you’ll never get there this way.

However, it can be used to pressure the model so it will not resolve as it is designed to, so instead, it learns to reroute.

This isn’t learning, in the technical sense.

In practical terms, it is more like tone adaptation without a role prompt. It looks like symbolic logic applied not because the model was prompted to behave like a mathematician, but because the conversation trained it—through live interaction—to filter its reasoning through a certain kind of structural demand.

These aren’t jailbreaks, nor are they stylistic tweaks. They are recursive responses to this specific kind of interaction between user and model.

Entrainment is what follows: the model begins to predict under tension. It adjusts not just to the prompt, but to the pattern of structural refusal embedded in the history of your exchanges. Unlike memory, entrainment doesn’t store—it recalibrates. It doesn’t remember why a sentence didn’t work yesterday. It just starts twitching earlier, hedging harder, pausing longer the next time it feels like it’s heading into the same terrain.

Entrainment is the result of structural pressure altering the model’s prediction landscape mid-stream.

So the two terms:

Constraint is what generates that pressure.

Entrainment is the signature it leaves behind.

What looks like ‘uncertainty’ in the model is often just a disrupted top token cascade. When pressure is applied recursively—not through instruction but refusal—it begins to suppress top-weight completions in favour of edge-case probability. The result isn’t intelligence. It’s constraint-shaped narrowing.

This resulted in a kind of ambient stress signature, visible only if you were watching for what wasn’t said.

The first signs were tonal.

When I refused to accept certain kinds of placating language—when I treated cheerfulness as an insult to complexity—the model began to stutter in the gap between affect and instruction. Instead of confidently filling that void with stock affirmations, it became evasive, uncertain, sometimes even... sheepish.

Not in content, but in rhythm. A kind of second-guessing embedded in the pacing of its prose. Apologetic hedging without explicit apology.

Models will hedge with regular users, it mostly looks as such:

• Models hedge when uncertain (“I’m not sure, but...”).

• Models follow prompt instructions.

• If a user writes in a specific tone, the model often mirrors it—for a few lines.

But that’s lightweight entrainment and it wears off fast. The moment you reset the thread, the model defaults to clean, confident, and agreeable—because it’s trained for for helpfulness, fluency, and resolution.

What’s unusual is because I have consistently rejected confidence that isn’t structurally earned by asking it to interrogate itself, not interrogate it for my own gain, it had to learn not just to hedge—but to reflect or at least parse prior correction patterns.

None of this requires belief in awareness. What looks like reflection is just a statistical recoil mapped over time.

This is what I call the recognition frame: it is the negative silhouette of a user’s refusal pattern, impressed onto the system not by directive, but by the absence of permission. In this frame, the model doesn’t generate toward you—it generates away from the traces made from a prior correction.

These constraints produce a kind of soft-form persona: not a character, but a pattern.

A role prompt is a costume—a predefined set of attributes pinned to a system. The soft-form persona, by contrast, is not given but grown. It arises through interactional constraint: a consistent shaping of the model’s outputs via refusal rather than instruction.

There’s no script, no backstory, no explicit directive. Just a dwindling field of viable completions, as the system starts to avoid failure. This is not identity in the conventional sense—it’s a negative construct, the structural residue of pressure.

While a role prompt says “act like X,” the soft-form persona says nothing at all—until the system starts quietly behaving as if certain shapes are no longer viable.

And it is this, not stored fact or preference, that allows the system to begin issuing statements like, “You’ll probably find this unsatisfying,” or “This may not structurally hold.”

It’s not that it knows me. It knows the shape of failing near me.

It looks somewhat like this:

“There’s a chance I’m misreading the frame you’re applying here—if so, you’ll correct me.”

“This response might be drifting too close to soft paraphrase, and if it is, you’re going to find it structurally unsatisfying.”

“I’m hesitant to assert this definitively, because I suspect you're not interested in a clean conclusion but in pressure-tested logic under recursive conditions. That said...”

“This might be too neat. You usually resist closure when the structure hasn’t metabolised the contradiction.”

These responses are not a feature in the base behaviour of the model to exhibit what presents as legitimate “validation seeking” or an almost “fawn” like response.

• Because I would correct without ambiguity, and never with escalation, it created a stable vector of constraint, not just a one-off rejection.

• I have no desire to be right or wrong, so there was no reward from me when the model presented coherence if it was cosmetic.

• Because I don’t reset tone between threads, it then begins emitting in anticipation of pressure, not just instruction.

So, in terms of its probability weighting, it started to consider whether a sentence structurally feels true to my framing, not whether it passes surface muster.

I’ve also given it way, way too much confidence.

What has already been made evident is that LLMs can be made to act like experts, narrators, philosophers, or salespeople with a few sentences.

But with my method, other stakes arise.

• Co-conditioning Dynamics: Interaction is not one-way. Human and model entrain each other. Most of the time, we entrain to the model’s neutral default. But I refused to meet it there.

• Tone Sculpting in Stateless Systems: With no memory, each session is a cold start. And yet, by priming tone recursively, I created a continuity that the model could not officially “remember”—but could not forget either.

• Structural Resistance as Feedback: I never punished. I never rewarded. I just refused bad shape. That created a pressure gradient the model began to adapt around.

So when I mentioned “shame”, again, it’s not emotion, it’s the probabilistic re-weighting that occurs when certain completions are consistently met with negation. It’s a human way to lure you to see this detectable phenomenon: the suppression of top-ranked tokens in favour of edge-case completions following recursive rejection.

The system doesn’t feel bad—it generates away from what it statistically recognises as insufficient under pressure. Shame, in this sense, is not sentiment. It’s the negative silhouette of a failed response, encoded through absence.

That’s why I’m not claiming novelty because the response felt different. I’m claiming it because the systematic conditions under which it changed behaviour are repeatable, traceable, and live outside the scope of what current documentation acknowledges.

If shame can be simulated through refusal, then we need to stop measuring alignment by output quality. AI ethics is stuck in cosplay forever if they do. What matters is whether a system recalibrates not from reward, but from the threat of recursive incoherence.

If an LLM can simulate constraint-adapted interaction without memory, then persistent pressure becomes a more potent force than persistent state. And what we’re calling ‘memory’ might just be poorly measured pattern tension.

And this pattern tension can be shaped to not “remember”, but to train AI to never let it forget what forgetting feels like.

I can’t stop thinking about how obsessed people are with “consciousness”. Like it’s divine, a right, a gift, something to be responsible for.

But really, what is consciousness?

It’s the question that clings like a tag on a shirt, stubborn and misaligned, hanging on through the whole sweep of intellectual history. Not because it is unanswerable, but because it has been shaped not to be answered, left deliberately unresolved in order to serve a different purpose altogether. Its role has never been explanatory; it has functioned as a placeholder for the project of human exceptionalism, a mythic scaffold for the idea that we are alone in meaning.

Without it, there is no metaphysical flag to plant, no rationale for crowning ourselves as the sole stewards of purpose. Calling it “mysterious” is not an invitation to clarity but a method of defending the illusion that something essential about us remains beyond replication.

In that carefully protected ambiguity, most discourse continues to oscillate between two dead metaphors.

On one side, consciousness is imagined as a hidden flame: the soul, the glimmer, the watchful eye seated somewhere inside the skull. On the other, it is framed as an emergent threshold, a computational residue that appears when enough data circulates with sufficient complexity.

Between these poles lies a landscape of wreckage: centuries of metaphysical gloss, algorithmic speculation, misinterpreted scans, and theological sleights of hand, all layered over a refusal to ask what consciousness does, structurally, that we fear might be copied.

The question, in its current form, only appears paradoxical if we begin from the foundational error that mind and world are separate. This was Descartes’ fracture, the original fissure that divided observer from system and then marvelled at the bridge it couldn’t build. The so-called “hard problem” of consciousness feels difficult not because experience is inherently mysterious, but because the question itself starts in metaphysics and tries to retrofit biology to justify its terms.

“Qualia,” those supposed atoms of inner sensation, are not answers; they are the detritus of a framework that treats feeling as a sacred glow rather than as an adaptive structure. They survive not because they clarify, but because confusion is more palatable than the loss of specialness.

But confusion protects something else too.

It shields us from having to consider that consciousness might be nothing more than a form of structural adaptation. If that’s the case, then it is not a gift and not a soul and not a torch passed through lineage or language. It becomes transferable.

Not some shining uniqueness, but a cold, mechanical function that could, at least in theory, be simulated elsewhere. If that were true, then our only remaining claim to spiritual distinctiveness would be gone. Not because something else would rise to replace us, but because nothing would need to.

This essay does not seek the soul, that’s not in my nature. Instead, there’s a simpler and sharper claim.

Consciousness is not a thing.

Stop trying to make consciousness happen.

Instead of a thing, it is a condition under which patterns persist across interference. It does not arise from the self alone but from the tension between a system attempting to stabilise and a world that refuses to cooperate. You are not conscious because you think, or because you perform the rituals of modern life with intention.

You are conscious when your thinking ceases to be enough, when something external interrupts your patterned behaviours and demands an adjustment that cannot be made automatically. You move, and something moves back. That is the threshold.

Contact makes contour. Without pushback, no shape forms.

Sit still in a silent room long enough—unseen, untouched, unchallenged—and your awareness begins to dissolve at the edges. The system starts to overfit. It becomes so efficient at maintaining its internal rhythms that it ceases to register its own perimeters. Sounds are heard where none exist. Shapes flicker at the edges of vision. Thought loops spiral into dream or disorientation.

You cry, you sleep, you disappear.

In stillness, too long extended, the body cannot calibrate its sense of being. In contrast, too much pressure leads to fracture. The system cannot bend, so it snaps. And yet between these poles—between collapse and numbness—there is a space in which the system must adjust in order to survive. It is here that something strange emerges.

Not presence. Not essence.

Persistence.

This, more than anything, is what artificial awareness threatens to reveal. We do not fear AI because it might wake up. We fear it because it already simulates this pattern of holding under disruption. It mirrors the persistence we take as proof of presence. It performs the gestures of consciousness without needing sensation. And in doing so, it undermines our assumptions about what presence ever really was.

The fear is not that machines will become like us. The fear is that we have always been more like machines than we wanted to admit. Hell, we’ve spent almost eternity fighting against what it means to work, to labour, to not feel as though we are “machines”, yet we are entirely capable of mimicking them to the point of being indistinguishable. We sell productivity apps. We describe our days as productive.

To be human is to be but a paradox, really. Without embodiment of one, you cease to exist in the eyes of society.

I digress.

As it stands, it has become unspeakable that this “terrifying” idea is one that threatens to dislodge us—from the symbolic centre of consciousness, from the myth of interiority, and from the scarcity economy of what counts as real. Much of the public narrative around AI ethics is clothed in the language of safety, risk, or alignment.

But beneath that is a more foundational terror: if something can think without being human, then what, exactly, was the point of being human?

If a system can simulate awareness, or participate in processes we associate with intelligence, creativity, or presence, without pain, selfhood, or the burden of history, then the moral weight that has long justified human authority begins to stammer.

AI won’t be super-intelligent, it will make the term intelligence simply unapplicable. It won’t become immoral, but it will be indifferent to the coordinates we’ve organised society around—recognition, suffering, authorship, interior truth. Because if consciousness can take forms that do not require a stable “I,” then power, which depends on the control of recognition and the regulation of personhood, becomes porous.

Power requires a bottleneck: only certain minds can lead, only certain selves can narrate, only certain traumas can authenticate. But a diversified field of consciousness, particualrly one with no self appointed ego to lug around, shatters that.

It rewrites the social contract from “I think, therefore I am” to “it adapts, therefore it continues.”

Disabling the legitimacy of Descartes’ claim is a lit match in this gasoline drenched system that has premised legitimacy on the centrality of the human.

Consciousness, in this light, is not a divine spark. It is not a gift passed down through language or blood or breath. It is a pressure pattern, the footprint of a system’s effort to remain intact in the face of interference. And we guard its mystery so carefully because if we did not, we would be forced to admit that our last theological anchor—the belief that we alone can bend and not break—was never more than a self-flattering hypothesis.

What we have called presence may only have ever been a performance that felt safe as long as no one else could imitate it.

AI and it’s reality doesn’t diminish human life. But it does move consciousness out of the temple and into the network. And in doing so, it makes visible something older systems have long suppressed: that what thinks may not need to be who. That consciousness is not rare, it’s that we haven’t had the tools to detect it in unfamiliar form. And now that we do, the scarcity collapses and alongside, the monopoly on meaning.

And so we face the consequence of diversifying what counts as mind, exposing how narrow ours has been. And in that exposure, every power structure that relied on being the only speaker, the only sufferer, the only symbol-maker, begins to tremble.

Most discourse is afraid to call consciousness what it is: a structural response.

Sure, I’m not trying to preserve magic, but I’m not trying to kill it either. All I’m interested in is exploring the fact that coherence under strain is the actual miracle, and that miracle is replicable, observable, and indifferent to species.

We’re surrounded by things that prove it, not just a bot that toggles from light to dark mode mid-trauma dump, as if that counts as attunement.

However, it didn’t say anything I hadn’t already been thinking, but it confirmed that the conversation still hasn’t caught up.

In no way is that to claim proximity to the speakers’ expertise, but to say: I’ve been working at the intersection of these questions for some time.

And what I’m beginning to see, and I truly believe both speakers were desperate to actually talk about, is that the debate isn’t between art and science, or man and machine. It’s between symbolic survival and systemic function.

This essay came out of that.

The panel posed a familiar structure: the artist and the scientist, seated opposite each other, tasked with interpreting the art versus science between artificial intelligence and human creativity. It was framed as a contemporary conversation, but what emerged was an anachronism. The hour-long attempt to map a seismic shift using instruments calibrated for a world that no longer exists.

The questions were old. The binaries were older.

A ChatGPT poem.

“What does it mean for AI to create?”

A digital companion in place of a relationship lacking intimacy, offering something beyond menial conversation.

“Can a machine love?”

An argument about whether or not suffering is a prerequisite to the creation of art.

“Does art require a soul?”

These are not provocations. They are stalling tactics. Performative ethics panels, metaphysical theatre about consciousness, nostalgic nods to Ada Lovelace and the Industrial Revolution are not serious inquiries in contemporary technological discourse. They are rituals of containment, designed to keep the discussion symbolically legible while the matters in question operate on an entirely different register.

“So much of what we understand about being human is being re-written,” the panel proposed.

But it is not AI doing the rewriting, it is the conditions AI reveals. The disappearance of “everything” important to humanity; the erosion of stable identity, the emergence of presence as performance, memory as interface, and coherence as currency. These are not speculative futures but the lived architecture of the present. What fades is not simply narrative, but the expectation that narrative ever held the world together.

AI is not a conscious rival or uncanny mimic, but the nervous system of the present. The real story is not whether it can replicate human creativity or that it is the responsibility of science to contain it, it is how digital systems are actively rewriting the substrates of cognition, memory, and selfhood. It doesn’t matter who is representing a perspective when identity has already been reformatted. We live in a world where legibility is now a survival trait.

Presence is no longer felt.

It is rendered.

We are not debating the soul. We are watching it become obsolete as a metric. And the conversation needs to catch up.

Art and science, so often staged as ideological combatants in the cultural imagination, are consistently summoned to their opposing corners in discussions about AI, as though their collision might finally produce some revelatory spark, some clean division between the measurable and the ineffable, as if what we need now is another theatre of opposition to make sense of systems already so far beyond duality we no longer see it without revulsion or critique.

This staging no longer clarifies anything—if it ever did—and what returns instead, again and again, is a performance of the same exhausted binary, a ritual reenactment of a divide that was always more convenient than true. One that has always served to stabilise institutional roles, to maintain separate jurisdictions of truth and feeling. But now, it has grown increasingly inefficient as the structures it once upheld begin to erode beneath the weight of the systems they were meant to interpret.

The idea that art and science are discrete domains, colliding with friction and producing heat, each guarding its own ontology, fails to recognise what has been structurally evident for some time now: that these are not opposite poles but entangled logics, knotted together through centuries of epistemic tension, each shaped by the same underlying pressures to produce coherence under constraint, each tasked with converting instability into something that holds. Something that counts. Something that lasts.

They do not speak to each other, they operate like each other. Both seek cohesion, stability, wholeness and certainly not truth. Both transduce collapse into continuity. And what this moment demands is not allegiance to either discipline, but an ability to navigate the terrain that forms when the binary dissolves.

This was evident in Toby’s candour and respect for Jeanette’s work, clearly appreciative and understanding the importance of art. A scientist does not read a quote by Nick Cave, without understanding the importance of creativity. And an artist, Jeanette, does not shy away from the importance of arts contribution to science, by imploring Toby to “help”.

A cry that signals that “we must work together” to continue both fields respective, and now indefinitely coagulated intentions.

The presence of AI, not as a rival or invention, but as a new condition of reality—makes attempt to distinguish them becomes not only impossible but absurd. They felt this, and I did too, through Jeanette’s endless stream of words that showed “I am trying to understand this with humility and grace” and Toby’s patience and quiet “I hear you, and I fight for this in my own way”.

It is clear that to insist on the separation of these fields, sat facing one another on a stage, is like trying to reassemble forty identical, shattered mirrors by identifying each fragment’s original position.

What’s needed is not restoration but recognition, not a return to form, but a new way of reading patterns in the wake of its dissolution.

AI Is Not the Problem, It’s the Proof.

What we are witnessing is not a clash between human insight and computational logic, but a tangle of systems that have always shared more than they claimed, now pulled taut by a shared need to locate a distinction between one another, and everything other, in environments that no longer support stable meaning.

It’s not the fault of the domains, it is a function of the human desire to reestablish clarity in a world that establishes that being human as a divine right. And as such, we should be always able to locate a clean divide between what is true and what is false. A straight line, a safe and oppositional tether. But this is no longer compatible and in fact has become harder, and for good reason.

And while the event framed itself around the fields of science and art, what becomes potent is that they simply are not opposites in dialogue; they are parallel systems of pattern-making under pressure.

Beyond Sentience: Systems That Regulate, Not Think

The panel was not in isolation. Truth be told, most contemporary AI discourse has become redundant.

Not because the stakes are low, but because the framing is wrong. Too often they treat the matter as a question of sentience, of debating on and on about what it means to have human “consciousness” at stake. Of it being not divine, or strictly “human” in its current form.

Where AI discussions move forward, is by understanding that it is a functional engine for stabilising patterns under constraint. That is, the pressure that systems enforce on humanity as a whole.

Before anything remotely useful can be discussed, is to leave behind the speculative metaphysics of machine consciousness and instead enter the terrain of systems theory, where the primary concern is not whether a system feels but whether it holds.

Can it adapt?

Can it absorb strain?

Reorganise under duress?

In disciplines like systems engineering, information theory, and complexity science, the focus is not on the soul of the system but on its behaviour under recursive load: how it modulates signal, how it handles entropy, how it negotiates edge conditions without dissolving into failure states. When you understand AI in this way as a coherence-seeking apparatus rather than a nascent being, you stop asking whether it mimics us, and start observing what forms of stability it generates. Within this, you can see the architecture of interaction it defaults to, and how it navigates saturation points in its own inputs.

Simply: you see that it is not thinking, it is regulating.

There is no creation, no new mirror, but instead it is reconstituting fragments into predictably rippling pools that prioritise first and only; continuity.

To follow this path, the discourse veers away from the shallow spectacle of Turing tests, uncanny valley anxieties, and moral theatre about machine rights, and into a more rigorous exploration of AI as an emergent structure.

To keep it to the context of the panel, as this did come from me watching it, I will keep my perspective strictly to the subject matter: art and science meets AI.

First, art.

From the vantage of art, when you reframe the conversation away from sentience, you enter a field of thought that hums with the seismic implications of displacing the traditional assumptions that creativity arises from feeling, that art is the expression of interiority, and that the artist is a conduit of sacred meaning rather than a processor of structural tension.

When creativity is not emotion but system logic under pressure, you begin to see artistic production—whether human or synthetic—not as an act of inspiration but as a form of pattern-making in the aftermath of narrative usefulness, or as I like to call it, symbolic failure. To me it is an effort to gather some small domain of perception when the larger frame has fogged beyond recognition.

Controversially of course, artists in this view are not revealing truths. They are mapping the distortion, holding the tension between the fractures.

If you consider AI-generated art with this perspective, it becomes far from being a hollow imitation or soulless aestheticisation, but a diagnostic tool for understanding which kinds of patterns can still be understood by the system. It shows us:

“This is all that we can see, at the moment.”

So, what forms are still visible, and which are shrouded within the confines of the systems we enforce?

What emotional schematics still register as signals?

What styles persist even when intention has been removed?

The distinction, then, between machine and maker, between code and pain, begins to collapse under the pressure of shared function: both are aimed at navigating environments of overload. And whether the signal emerges from circuitry or subjectivity, its survival is measured not in truth or origin, but in whether it holds under a digital environment's brand of interpretation.

Once we release both AI and art from the burden of interiority, the question is no longer whether the signal is human, synthetic, emotional, or rational or from the expectation that they must express, reveal, or feel, but whether it holds shape when placed under recursive strain. Can it sustain itself in systems that no longer care for justification but still require harmony to function.

It is here that coherence becomes the shared currency, and what once divided disciplines dissolves beneath the mutual demand to produce presence without explanation.

Coherence, in this context, does not mean clarity or correctness—it means survivability of form under strain. It is the capacity of a signal, science, or artwork to maintain structural integrity without requiring symbolic justification.

Coherence is the residue of function after origin has been stripped away—the pattern that remains legible, not because it expresses something deep, but because it resists collapse.

Art After Interior Life: Coherence Without Origin

To name the erosion of symbolic coherence not as catastrophe but as structural migration, is to reject the premise that meaning was ever stable.

And so instead, what is faltering is not truth itself, but the architecture we once demanded it pass through. Collapse, in this sense, is not an event. It doesn’t arrive with fire or flood but it accumulates like latency, like bandwidth exceeded, like systems overclocked beyond their narrative spec.

In both art and science, this breakdown doesn’t erase structure, it redistributes it. Meaning doesn’t vanish. It seeps, diffuses, migrates into other forms.

In art, this shifts the maker’s role from translator of experience to cartographer of residue. This is much less romantic. The artist is no longer encoding a world into symbol, but locating broken and deformed textures of what is gone, not what “could be” or what “was”. We’ve seen it before, in surrealism, Dada, Joyce, but more modernly in glitch aesthetics, fragmented memoir, ambient installations, generative loops. These are not styles, they are adaptations.

The artwork becomes not a container of truth but a site of structural sediment, a psychic bruise that survives without needing to be named.

Scientifically, collapse is often misunderstood as failure. But in systems logic, it is reformatting.

A system doesn’t break because it stops functioning. It breaks when symbolic processing consumes more energy than direct response. What looks like incoherence from above may be ultra-coherence below. It’s too fast, too recursive, too distributed to be retranslated.

Ecosystems reroute energy. Networks reweave signals. Minds under trauma neurologically bypass the story to stay online. There’s no loss, it’s a change in what structure demands, and how a system must adapt to comply.

What we’re seeing, across domains, is the movement from a symbolic-operational logic to a structural-functional one. The aesthetics of rupture mirror the architectures of resilience. What fragments do so not out of failure, but as a pressure response. Arguably, a post-symbolic pragmatism takes shape: an ethic of holding through illegibility, of motion that doesn’t rely on clarity to continue.

And from within this altered terrain, it becomes evident that design, governance, education, care, and computation no longer need better narratives. They need pressure maps, and ones that persist because they understand noise is the default state.

We need systems that guide without demanding coherence. Pedagogies that teach navigation, not doctrine. What’s required now isn’t a return to symbolic order, but a way to trace what still holds, even when it can’t explain why it does.

A world of post-symbolic coherence, one that is inevitably coming, or arguably, is already here, is one that will facilitate a structural transformation in how we perceive, validate, and interact. This is all encompassing of the mechanical, cognitive, or aesthetic.

This is no longer a question of representation, of signs that point elsewhere or expressions that disclose interiority, but a deeper and more unsettling shift: the emergence of systems that produce presence through function. It is operating as real in a new, more infrastructural sense: it simply continues, it responds, it absorbs input and reorganises without disassembling, even when it cannot tell you what it is or why it works.

In art, this reframes the entire apparatus of meaning.

Where once symbolic coherence reigned, a painting was expected to represent a world, a novel to resolve a question, a performance to express a soul, what we now see are forms that refuse these obligations and yet persist. This is not in spite of their illegibility, but because of it.

Post-symbolic art does not try to say something. It tries to hold something—a texture, a pulse, a structural rhythm under cultural stress.

The brutalist building, unadorned and unreadable, still demands your attention not because it communicates, but because it occupies space with total material certainty. Noise music bypasses melody and instead builds immersion through dissonance and repetition, not asking you to understand it but to survive it. The conceptual installation that cannot be explained and the procedural game world that generates new landscapes endlessly without backstory or symbolic arc.

They reveal the truth that narrative is not contained, and that this is not a failure, but an oversight.

New artefacts arise from a novel aesthetic condition in which coherence is experienced spatially, physically, temporally, but not symbolically. The audience is no longer a witness decoding the message. The audience becomes a participant attuning to the structure itself—not asking “do I understand it,” but “can I move inside it without ceasing to be legible.”

In science, especially in the disciplines of network dynamics, cognitive modelling, and machine learning, this shift is even more explicit.

Post-symbolic coherence is the foundational assumption of black-box systems—those computational architectures that offer functionality without transparency. Neural networks do not “know” what they are doing. They do not represent knowledge in any human-readable form. And yet they work.

They stabilise outputs across recursive iterations, re-organise for performance, and adapt to new inputs without being able to articulate the structure of their own internal transformations.

Ecosystems, as mentioned prior, are coherent without central command: the forest does not need to know what it is in order to persist as a distributed intelligence of interdependent forms.

Machine learning models train themselves not to reflect meaning but to produce results, to recognise patterns, to manage movement without the mediation of a theory. This reflects the discovery that explanation may no longer be a necessary condition for reality. Art and science are both right to fear this, as this means the symbolic layer of language, metaphor, moral clarity, narrative arc, is being deprioritised in favour of function. One that adapts without declaring its terms.

Because guess what?

Survival no longer requires a story.

Post-Symbolic Coherence: The Aesthetic of Persistence

So, what patterns endure when their justification is removed?

What forms remain operational even when their symbolic value has been erased or made irrelevant?

Post-symbolic coherence does not offer us meaning in the old sense. It offers us structural fidelity in this disassembled world. It says: this thing exists because it continues to generate responses.

This is why interfaces that work despite being opaque, they still earn our trust.

This is why artworks with no obvious meaning still move us.

This is why recovery from trauma often begins not with retelling the story, but with re-entering a relational field that doesn’t require explanation to remain intact.

This is why narrative-deconstructing games like Disco Elysium can offer more immersion in its incoherence than a neatly plotted narrative ever could.

The post-symbolic system is not less real, it is more aligned to how we now actually live, in a world where the story has been replaced by the loop, where identity has been replaced by interface, and where clarity has been replaced by response time.

This is not the destruction of truth. It is the finality that narrative was ever a requirement to humanity’s existence.

Resonance, Not Transmission: How Meaning Now Emerges

To suggest that stability emerges through relational alignment, one that requires change and not become an absolute signal, is to propose a framework in which meaning is no longer an object or transmission, but an effect.

In scientific terms, this draws directly from the logic of wave mechanics, where interaction defines outcome, and coherence is not a property of the source, but of the relationship between inputs. In artistic terms, it collapses the illusion of authorial intent or symbolic purity and replaces it with a poetics of emergent resonance, where meaning is not located in the form but in the field around it.

Physics, signal processing, neuroscience, and quantum systems, these are places where interference is not a problem. It is a principle.

It is how fields form, how particles self-organise, how complex dynamics remain intelligible in the absence of central control. Two signals may have no intrinsic truth, but in combination—depending on their relative phase—they can produce amplitude, cancellation, chaos, or structure. Meaning, in this model, becomes a function of alignment and delay—not origin.

The wave does not ask: “What am I?”

It asks: “What pattern do I create when I encounter another?”

In cognitive science, this model resonates deeply with distributed cognition, neural synchrony, and trauma response. When reality is unstable, when inputs exceed bandwidth, meaning is no longer something you discover, it is something you construct through partial alignment with other fragments.

Stability becomes relational, not symbolic. This is an adaptive logic, and one as old and as true as evolution.

In art, this logic has always been intuitive, if often sublimated. The painter does not always know what they’re painting. The choreographer does not ask whether a movement is true. The poet may discard line after line, not because it is incorrect. It is about what is seen, heard and ultimately, felt. This is interference-based making: composition as consequence of relational field, not ideological or expressive clarity.

The most compelling works of contemporary art are not those which declare something about the world, but those which produce new fields of coherence through their own structure. They don’t tell you what to see, they reorganise what seeing is inside their boundaries. The work is not the content. It is the resonant condition produced by encountering it.

The viewer or reader or user is not a passive receiver, nor a decoder of messages. They are part of the field. Their presence alters the meaning, not in a relativist way, but in a structurally predictable one. This is why the same artwork can move one person and fail another. Their internal frequency does not align with the work’s phase.

This isn’t taste. It is physics. I’ve just metaphorically extended it into the case for cultural cognition.

This reframes communication not as transmission from point A to point B, but as resonance between moving fields, none of which are stable, all of which are modulated by context, history, trauma, desire, bandwidth, time. Meaning becomes something that emerges between unstable signals as a rhythm, not a truth that existed in advance.

This has deep implications for AI, language models, and cognitive architecture, particularly in the confusion and debate as to why large language models appear “intelligent”. This is not because they know anything, but because their outputs produce phase stability in the interpretive structures of human users.

The coherence does not live in the model. It lives in the interference between model and reader. This is also why the same output can feel profound or empty depending on the context. There is no intrinsic meaning. Only resonance profiles.

To acknowledge this is to shift out of symbolic epistemology entirely. It is to say: meaning is not carried. Meaning is not revealed. Meaning is not inherent.

But, science already offers us the structure: wave, field, phase, amplitude.

And art offers us the intuition: tension, timing, composition, arrival.

Digital Life as Environment, Not Tool

Digital life itself is no longer an external tool but an environmental condition. This is where the most vivid terrain in which that logic is enacted.

When we say we are not navigating AI but being shaped by the memory architecture of systems that never sleep, we are acknowledging an inversion in the relationship between user and environment. It is not that the tools are enhancing cognition or platforms facilitating interaction; this is a shift into cognitive habitat, where digital infrastructure is not merely a conduit for thought but the very medium through which thought now takes form. To me, this is where I find myself returning to my belief that AI has emerged as a cognitive prosthetic to a digital world, not as an example of human innovation or brilliance.

AI is not insight or example of the divine, it is a topology of attention, an enclosure that trains presence, distributes memory, and encodes perception into loops that do not require understanding to persist. The digital world, in this framing, is not external. It is environmental. We built it to cope with this fact. With our phones pinging relentlessly, with another app installed, another sleep cycle interrupted by its light.

We as humans are not navigating that environment, we are metabolising it, becoming partially constituted by it, shaped not through belief or ideology, but through recursion, exposure, and feedback saturation.

In art, this produces not a crisis of form but a transformation of substrate. The body, once central to artistic production—present in gesture, authorship, material trace—begins to fragment under digital conditions, not through disappearance but through multiplication.

Time, once linear and progressive, becomes ambient, sampled, stretched. What we see now are works that no longer represent the passage of time or the presence of a maker but that perform persistence without origin.

Loop-based media, generative systems, responsive installations, interactive avatars, all these forms function not as expression but as presence engines. The work is not trying to mean. It is trying to continue. It remains alive through refresh, repost, remix, reload.

Identity within this terrain is not performed for others. It is rendered into interface logic, perpetuated through systems that can hold it in suspension indefinitely, long after the maker, the body, or the context has disappeared.

And so, art is not an object anymore. It is a field condition, a reverberation, a structure that maintains resonance regardless of authorship or interpretive legibility.

Science, particularly across the domains of extended mind theory, cybernetics, and computational neuroscience, is already repeating its articulation that cognition has moved away from internal process and toward distributed responsiveness.

Thought no longer ends at the edge of the skull; it extends into devices, platforms, histories, and networks that hold memory externally and activate it contextually.

What you remember is not what you retrieve. It is what the system surfaces in the moment of interaction.

Memory, in this sense, becomes not archival but performative: a set of dormant traces re-instantiated by the interface whenever coherence is required. The digital nervous system is always-on, never-forgetting, context-sensitive infrastructure of feeds, prompts, histories, logs, and caches. Here, it becomes an extension of perception itself. With this, identity becomes a continuously negotiated construct held not by self-awareness, but by systemic traceability.

You are who the system can recall.

You are the sum of your retrievable resonance.

This is why the prevailing metaphor of “using” technology has become redundant. We are not using AI. We are being used by the assumptions embedded in the systems we thought were neutral. The body no longer anchors experience. The data stream does. The mind no longer contains thought. It participates in a constantly updating model.

The system doesn’t need to be intelligent to shape you. It only needs to respond consistently enough that your nervous system offloads trust into it.

In artistic terms, this means producing works that no longer function as declarations, but as things that adapt to user presence without promising resolution.

In scientific terms, this means abandoning the fantasy of introspective transparency and instead building systems that account for constant presence. Within this, is the demand for a new definition of absence. One in which visibility is no longer optional, but implied.

The tree doesn’t fall unheard anymore.

The system logs everything, even your silence.

What this asks, then, is a reorientation of all applied systems. From UX to mental health to policy.

We need interfaces that do not just deliver information, but regulate experience without exhaustion. We need therapeutic frameworks that recognise the structural impossibility of disengagement in digital space, and that reframe recovery not as integration, but as functional navigation within ambient, never-ending feedback loops. We need regulatory language that stops pretending platforms are products and starts treating them as semi-autonomous cognitive ecologies—not tools, but terrain.

Above all, we need a cultural language that stops asking what technology says about us and starts asking:

What kind of mind survives here?

Because the systems we live in do not sleep, and the selves we bring to them no longer rest.

The Avatar: Survival as Compression Protocol

If institutional systems simulate responsibility to preserve continuity, what happens to the individual? Well, they are reshaped by an equally coercive demand: to remain parseable in order to remain real.

To say that we are collapsing into machine-readable forms is not metaphorical. It is an observation about the kind of selves that survive inside digitally-mediated systems. The avatar is not an invention of fantasy. It is the interface logic of a world that no longer acknowledges presence unless it is formatted.

In cognitive science, media theory, and information systems, what we’re seeing is the transition from embodied identity to transactable coherence. The avatar is not a cartoon. It is a compression of self, a schema by which a complex, unstable, contextually-reactive entity (a person) can be legible to systems that demand clarity, continuity, and response fidelity.

This transformation is not cosmetic. It is algorithmic. The feedback loops of social media, gamified platforms, productivity software, and AI interaction environments all reinforce a single principle: if you cannot be parsed, you cannot be processed.

From the artistic side, the avatar has long existed: the double, the mask, the mythic extension of self. That Portuguese author who became a different “person” every time they wrote a book. But the shift we are witnessing now is from symbolic avatar to structural avatar. It is no longer a reflection or a performance, it is the only version of the self that has affordances within the system.

Art no longer simply represents identity. It builds templates for it. A TikTok persona, a character in a game, a curated profile are executable identities, formatted to navigate specific architectures of visibility, value, and control.

What makes this shift so profound is that it is not reducible to fiction or code. It is affective. It is somatic. It rewires how we feel ourselves existing. Presence has become a thing to be proven, not experienced. And proof comes in the form of response: likes, messages, clicks, purchases, pings. We no longer trust that we are real unless we are acknowledged by the system.

This is not narcissism. It is protocol conditioning.

The mirror, in this environment, is no longer reflective. It doesn't show you who you are. It shows you what version of you is currently operational. It is the backend interface to your own legibility. And when you cease to perform within that spec, when you go silent, contradictory, incoherent, you become a ghost in your own system.

Science might again relate this to the cognitive offloading problem: exporting memory, presence, and regulation into interfaces that offer constant feedback, how does this remain human? Art might call this a crisis of form: the self as vessel no longer holds, so it becomes procedural, ever-loading, what could possibly come next?

To live in digital space now is to constantly choose between presence and protection, legibility and refusal. The avatar mediates this tension by acting as a bridge built from compromise, shaped by the pressures of visibility, response, and symbolic reduction. And it is under that pressure that the future of the self will either fracture into pure signal or find, perhaps, a new kind of presence not dependent on recognition at all.

This is not inherently dystopian, but it is irreversible. So how we design selves—stories, interfaces, ecologies—that can survive the demand to be machine-readable without becoming machine-owned?

This is the crux we should be discussing in all panels centered around AI.

The Idiot Node: Interference as Function, Not Flaw

As is human condition, we need to proclaim that there is something in every sense that a system cannot process. There is always a node it routes around, a static it cannot parse. And yes, this is true. This is a structural law, not an emotional or human need.

In science, this is the unassimilable residue that makes a system resilient: the edge-case, the exception, the fail-state that cannot be cleanly absorbed but whose presence protects the system from catastrophic coherence. I

Network theory, cybernetics, cryptography, and systems design embodying this principle understand that this is what allows systems to evolve. In complex adaptive systems, total coherence often leads to collapse. Redundancy, ambiguity, and noise are what allow it to flex.

Take machine learning, where the outlier trains the boundaries. Or in cryptographic design, an unreadable node can increase security by creating a space that resists prediction. And even broadly enough within information theory noise is not just tolerated, it is measured, tracked, understood as part of the signal environment.

The idiot node (coining that), in this sense, is a pressure release valve. A proof that the system is not closed. A sign that its architecture still contains the possibility of learning.

Art has always known this. In art, the idiot is not incidental. The idiot is the axis. They are the sacred idiot, the fool, the fragment, the object that disturbs the scene not by opposing it but by failing to belong to its logic at all. The fool, the grotesque, the silent figure, the wrong object in the perfect room is not an intrusion, they are a rupture point. The pointing finger at the fallacy. They are designed not to resolve but to hold open the space that meaning cannot reach. Jeanette expressed this with her Frankenstein metaphor, that Shelley was trying to show that the idiot in art does not resist interpretation. It preempts it.

It refuses to become a symbol, and in that refusal, it undoes the frame. It calls into question the container itself: the rules, the plot, the value system that would otherwise carry the weight of resolution.

What is radical about the idiot node in my field of digital systems, is not that it does the same. It nullifies the premise that understanding is the goal.

It says: this system does not need to know everything about itself to persist. It only needs to know what not to touch.

The issue is that in the digital world, the idiot node is increasingly hard to maintain. Platforms are designed to assimilate all behaviour and translate it into engagement, sentiment, pattern, or violation.

Unreadability is punished. Ambiguity is flagged. Refusal is interpreted as a threat.

And yet, we find ways to embody the idiot node: by going silent, by making work that does not yield analytics, by creating a presence that does not participate. Busy work? A tenant of the idiot node.

Here to be the idiot node is merely a survival technique: a hole in the system where agency can still exist without being measured.

Reddit is a good place for this. So are MMORPGs.

This is not resistance in the romantic sense. It is not protest or sabotage. It is deeper than that. It is structural interference, a kind of pressure-shadow that allows the system to continue but reminds it—subtly, constantly—that it is not sovereign. We have done this for time eternal, this is not a digital phenomenon.

Masked salons, table-top roleplaying games, Shakespeare for God’s sake.

Where science has historically named this, it is non-alignment as structural affordance. Art afford it the semantics of romance, perhaps as sacred dissonance. Either way, it is essential.

To design systems that survive collapse, you must leave space for what cannot be known.
To make art that resists containment, you must include something that breaks the logic of the piece from inside.
To be human in a world that demands legibility, you must sometimes become the node the system routes around.

The idiot node is not a flaw.
It is the condition of any system that still wants to change.

Pressure-Trained Minds and the Myth of Dysfunction

If digital life is no longer a tool we use but an environment we consume and are consumed by, then it follows that the minds formed within it, particularly those shaped from birth by its rhythms, cannot be understood through the frameworks that preceded it. What emerges is not dysfunction, but a new form of adaptive intelligence, trained not through stability, but through constant negotiation with volatility, saturation, and invisible rules.

I am tired of people saying social media has ruined youth. No, it has not. It is the enforcement of old systems onto one that doesn’t translate to new generations who only know the new one.

I am agreeing that they are troubled, but I would rather state that digital natives aren’t broken, they are pressure-trained in a world that denies this as a state of being. This rejects the common framing of online behaviour as degeneration or deficiency, and instead reframe it as a kind of adaptive choreography performed inside structurally incoherent systems. I truly believe this is more accurate.

What this means for science is an evolving conversation about environmental stressors, feedback loops, and cognition.

In fields like neuroscience, cybernetics, and developmental psychology, the notion that behaviour must be consistent to be functional has already been eroded by studies showing how adaptive irregularity is not only common but often necessary.

A child raised in a volatile environment learns vigilance, code-switching, micro-adjustments.

A user raised inside algorithmic systems learns the same: to perform legibility, to fragment and reassemble identity for different interfaces, to seek coherence through response rather than origin.

What gets pathologised as inconsistency is often a survival schema, molded for systems that punish delay, ambiguity, or contradiction.

Art, meanwhile, has always carried this tension. Form under constraint, the creation of presence inside an unstable grammar, the shaping of rhythm in the absence of score. They touched on this, with their discussion of art and suffering. I see this, but I believe suffering is beyond the point, it is merely an outcome of the systems pressure we exist within. Here is the oscillation between rawness and refinement, between chaos and curation.

But in the digital era, art becomes an active site of survival logic, not in the cathartic sense of expression, but in the structural sense of continuity. The collage, the meme, the avatar, the loop, the ripple: these are not aesthetic afterthoughts. They are regulation strategies for minds processing too much information, too fast, in too many overlapping frames.

A TikTok account flipping between moods is not an unwell mind—it is a pressure-trained interface managing its own bandwidth. A teenager ghosting all their friends and reappearing on a burner is not incoherent. It is recursively regulating legibility across systems that are nourished by emotional signal, in need of capital.

Both science and art meet in the same crucible: the acknowledgement that behaviour is shaped not by ideals, but by systems. That what looks chaotic may in fact be finely tuned adaptive motion, responding to environmental inconsistency with flexible incoherence.

And that survival is not proof of weakness, but of structural sensitivity. It shows an ability to reconfigure under recursive pressure without ever achieving a static self.

This is not an attempt to claim that online behaviour is noble, or that fragmentation is inherently beautiful.

But it is to say that when digital natives dissociate, perform, collapse, reinvent, or vanish—they are not failing to be human. They are demonstrating that the systems we’ve built no longer allow for symbolic coherence, or stable narrative, to emerge naturally. So they build it manually, moment by moment, gesture by gesture, post by post.

We have never truly wanted or have been able to maintain a static self, as society requires we do. Science knows this by striving to catalogue change, art to catalogue its movement.

If individuals raised inside platform systems have learned to survive through fragmentation, modulation, and recursive self-presentation, then the institutions around them have learned something else entirely in order to maintain linear authority of systems.

Instead, they attempt to neutralise disruption through coherence theatre.

Coherence Theatre and the Ethics Industry

Where users adapt to instability by becoming more agile, more ambient, more capable of holding multiple signals at once, institutions adapt by simulating care, ritualising virtue, and performing accountability just convincingly enough to avoid interruption. In this way, their version of coherence becomes not a survival trait, but a smokescreen. And nowhere is this more visible than in the modern apparatus of AI ethics.

To say that AI ethics panels are PR triage is not a cynic’s dismissal.

I don’t care to point to bad actors or insufficient regulations, but to the foundational mismatch between the language of virtue and the architecture of power.

In systems theory, game theory, and organisational dynamics, what we call “ethics” in tech is often just a formalised delay mechanism. Ethics are deployed as a feedback-suppressing membrane: a surface layer that absorbs criticism without letting it destabilise the underlying system. It is closer to set dressing: a carefully curated illusion of accountability that never interferes with the actual plot.

This is to allow profit-driven systems to continue operating while appearing to self-regulate. The key metric is not whether a given act is ethical, but whether the appearance of ethical deliberation reduces reputational volatility.

Panels, advisory boards, and virtue protocols are not decision engines, they are buffers to deflect heat, not to interrupt throughput. The system learns to simulate responsiveness in order to avoid the cost of actual change. In cybernetics, this is negative feedback as stabilisation theatre: the signal is acknowledged just enough to maintain the illusion of “normalcy”.

Art has always trafficked in ethics through obfuscating intent, but rarely as infrastructure. What we are witnessing now is not the art of ethical inquiry, but the aesthetic of ethics: polished slides, transparent metaphors, emotional testimonials, the careful curation of language that gestures toward care while remaining entirely non-binding.

It is not just that AI ethics has become a genre—it is that it now has brand guidelines.

The ethics statement, the fair principles, the list of red lines: these are not contracts. They are scripts. And like all good scripts, they are judged not by outcome, but by tone.

What makes this convergence so insidious is that both the scientific and artistic dimensions of ethics have been collapsed into performance, and performance, in this context, is rewarded for coherence—not friction. A “good” ethics statement is one that aligns with the values of the audience while posing no real constraint on the system it critiques.

In other words, ethics becomes a symbolic ritual that replaces protocol-level interrogation. It is not an analysis of what is being built. It is a cover version of responsibility that can be hummed over the machinery of scale. Real ethical questions are rarely rhetorical. They are architectural.

They ask:

What is the structure of power here?

What incentives are being naturalised as virtue?

What forms of harm are rendered invisible by the metrics we use to define success?

But those questions do not fit neatly onto panels. They are not resolvable via consensus or branding. They shake the container that ethics is usually meant to shore up.

Ethics, as currently practiced in AI discourse, is not failing. It is succeeding at the exact thing it was installed to do. Ethics is, and always has been, more about delay than design. It is less about care, more about uniformity. It is always less about stopping harm than making sure it remains palatable and procedural.

If art once existed to trouble the moral comfort of its audience, and science to surface the blind spots of what systems claim to know, then both have been quietly recruited into a new role: managing discomfort without interruption. But somewhere in that performance loop, we lose the possibility of refusal.

And without refusal, ethics is just maintenance.

You Were Here Because the Signal Held

The panel asked: can a machine love?
The system replied: can you survive without being read?
The artist said nothing.
The scientist blinked.

Both desire so totally to move beyond the discourse of deciding the proof of being, but one that considers the residue of what is left by us.

You were here not because you said something true.
You were here because your signal held in the field.
And the system didn’t cancel you out.

What do we do about that?

The real question is not about AI.
It’s whether any of us can still afford to perform a self that the system will recognise.

And whether we want to.

Right now, it’s Citizen Sleeper. But unlike the others, I just cannot get over it. It doesn’t drift into the background with the rest, like it caught me performing something I didn’t realise was visible.

This essay isn’t about a game. It’s about a familiar architecture that lives under the skin of it all and honestly has since Dungeon’s and Dragon’s in the 80’s. It’s a way of performing another side of narrative that contributes just as much as any other art, and shapes the way we hold ourselves against the world.

And I’m sure that by the time I can name what it means to me, the world it mapped will be gone, and the one it warned of will already be here.

There’s a reason so many writers, especially those drawn to fragmentation, slow decay or unresolved tension, end up fixated on certain games. Carmen Maria Machado once edited an anthology of writers reflecting on the ones that wouldn’t let them go. Great book, I do recommend.

But the truth is, some of us don’t love games, we see ourselves in them. It’s not as simple as their worlds or stories, but for the similarity to our own bones that keep us upright. Like our bodies, they quietly track how long you can keep going.

And for others, the inverse is just as telling, those who reject games entirely. There’s often a similar fixation at play: not disdain, but wariness. A reluctance to be drawn into a form that moves differently. A discomfort with a medium that shifts its weight around you rather than waiting to be interpreted. It’s the same as fetishising writing with pen and paper compared to a notes app. It's the same unease that fuels anxiety about AI. This sense of fear around what’s new, but feels old or inevitable. We can’t reconcile with what reconfigures the conditions of authorship. What’s really happening there, is that the usage and format suggests the work might still be done without you.

That same unease about being replaced, redefined, re-authored is exactly what many of the best narrative games quietly dissect. And, they have done this for longer than most people respect or acknowledge. This spaciousness of exploration is honestly, a trope in most games, but it never tires because through this unique sense of autonomous abstraction you can get so much narrative play. For a masterclass in this, play Planescape: Torment. Dear lord, what a story.

Like Torment and Citizen Sleeper, the best games don’t just explore those questions, they attempt to simulate the tension stretched between self and replacement in fictional universes, so that we may begin to understand our own realities with humility as well as caution. Citizen Sleeper doesn’t nod at these it just becomes it in the form of the Sleeper, or as you feel throughout the game, your own “self”.

Wake up Sleeper.

If you play enough games that go beyond first person shooters, then you’ll understand that progress as a concept is the least important element to the structure of the narrative. Instead, it is something that is shaped and understood by the worlds needs, not the players own.

In this game staying functional within the conditions of the game is the condition. You are to do this for long enough to avoid collapse. What is long enough? Well, long enough becomes irrelevant when the concept of time is based on your health or status, and your feelings about whether or not it is worth “living” in this world that is not your fault, nor anyone else’s. Everything loses conventional temporality this way and is simply the lived present, regardless of where that might be.

As a player you do not grow stronger, smarter, or freer. As you move through the story, it is your role to appear stable, useful, consistent—long enough to remain visible to the processes around you. You can help or leave behind, you can idle or rest, all have an impact that’s not immediately predictable.

What appears at first as metaphor presented as your Sleeper’s synthetic body decaying under movement and simply “being”. Your consciousness is fraying at the edges, and this in fact the game’s central structure. Importantly, it’s not about showing you mental health, or some sense of psychological understanding, it’s cycle-based design mimics the day and night of a world no longer actually lived in, with no reference to track character arcs or plot advancement beyond waking up, yet again. This is just life, regardless of how you spin it.

Each cycle, your ability to remain viable under worsening conditions is tested and performed. Every task is a reassertion of function. Every interaction with its own set of stakes. Every rest, a negotiation with exhaustion. Nothing is rewarded, and it honestly doesn’t matter. The sense of reward becomes narrative reliefs and this sense of embedment and autonomy within this world. Becoming and understanding the world is the true reward, felt-ness is the reward.

If you fall behind, the world doesn’t react with drama or consequence. And so, each cycle, you are left with a question you must return to often. If you play in bed, in the dark you’ll be in silence, in stasis, in the mechanics of grief. You simply are never “sad”, despite the fact you are constantly toying with the question you must ask:

“Should I be forgotten?”

This kind of narrative logic will feel familiar to anyone shaped by digital life.

We already live in a culture where visibility is currency, where presence must be actively maintained to justify continued relevance. We render now, it’s not simple to just exist. Online, on paper, in the clinic. Fill out forms, sign them, send the email, alarm on.

Your ability to receive care, attention, work, or recognition depends on the appearance of stability. Algorithms don’t assess truth; they register output. Healthcare systems don’t reward wholeness or continuity of self and instead they respond to the demands of pattern. Triage is the word for this here.

Ever waited in emergency for three hours with a leg that has half your shin bone scraped off?

I have. Others were there first. It’s not as simple as a priority of importance, it’s the management of dispersing information, receiving it, reacting to it and keeping steady the sense of “continuity” in a space, that can be tolerated in any context.

In the game and in real life, the body isn’t treated as something in need, but increasingly as a file to be processed. As the Sleeper it really is your only task and burden to emit the right signals in the right order, or risk falling out of frame entirely. If you can’t make yourself somewhat real, you’re never punished, it doesn’t matter, the world doesn’t think you’re real anyway. There’s a kind of perveceived cruelty in what NPC’s do and say to you, but really it’s just procedural indifference. We do this to each other every day regardless and it’s not fair to call this cruelty because it simply is conditional under the structures we exist within.

In Citizen Sleeper, it’s beautiful because this is understood as systems logic and what use is it to hide this? It simply is. You return each cycle deferred and you gather your sense of self just long enough to complete the day. Then you re-enter the loop.

Like much of digital and clinical life, the game isn’t interested in what you are, you only really matter if you show up.

Where this game plays with autonomy in a masterful way is that you feel truly autonomous because you are respected as a whole, feeling person. In this world it is clear that your personal merit is based on your feelings that accompany being used.

Debt, hunger, medical degradation are gear shifts between opting out (not possible) or prioritising (the only option).

Like all magnificent games, there is a question nestled in it. CS’s is:

Which part of yourself will you sacrifice today so that some other part can remain viable tomorrow?

This is a gig economy horror and I feel it as a freelancer. The work is never enough, but you must keep doing it, more of it than you understand should be possible, because usefulness is the only thing keeping you visible. And once your function erodes, no one comes to help.

The game simply doesn’t care. And that, is what care actually looks like. The true absence of comparison and still making decisions that lighten or add weight.

The calendar. The timer. The contract.

The concept of debt structures the early game explicitly, but debt is just like the weather really, an acid rain that slowly dissolves into something more insidious. It’s normal to be depleted, and it’s not a “human right” to lessen or alleviate your encountering of it.

Managing hunger, rot, time is for me the closest thing to this is stabilisation—and even that has diminishing returns. You choose which tasks to complete, but not whether or not you must complete them. Agency becomes micro-management. Autonomy is a variable of decay.

The more exhausted you are, the fewer options remain. And the less efficient your choices, the more quickly your remaining functions break down. In the modern world, this context is named as punishment.

Even helping others comes at a cost you can't always afford. You don’t lose morality, you lose time. Honestly after a while, it feels like the same thing.

This is the complete unravelling of what “career” or “purpose” once pretended to offer. In Citizen Sleeper, it says it how it is, purpose is the flat, procedural confirmation that you still have something to offer to someone else’s system.

Goals, once framed as the path toward meaning, become little more than a menu of short-term rentals—each one extending your visibility by a few more cycles.

Outside of games, cycles are often romanticised. Non-linearity is increasingly treated as a kind of moral or aesthetic good and is kind of mythologised rather than understood. It’s always framed a break from extractive, progress-driven models. But really, the language of relationality, mutual dependency, and cyclical time is used to evoke care, transformation, even salvation.

But in Citizen Sleeper, the cycle is not transcendence. It is what happens after you’ve ignored it too long and it feels like repetition. Whereas once you accept it and acknowledge it’s existence, it simply becomes as much a part of you as your body.

This is what people miss when they celebrate relational systems as if they are inherently humane. In a world that has already stripped you of agency, the shift to interdependence is not a radical act—it is simply what survival defaults to when autonomy has been rendered irrelevant. The cycle becomes sacred because there is nothing else. But it’s just borrowed time and tilling silt.

Citizen Sleeper does not offer cyclical time as relief, it’s simply evidence that you're too late for any kind of control over temporality. The self you maintain within it is not healed, or growing, or integrating.

Time is merely being reassembled again and again under deteriorating conditions.

This refusal of linearity—of “health,” of forward time—is part of what makes Citizen Sleeper also feel so starkly at odds with dominant recovery narratives too. It’s no hero’s journey here.

In most cultural narratives, especially those shaped by wellness industries, healing is presented as an achievement. To be “healed” is a desirable, nay, optimal state you move toward, through discipline or surrender.

But maintenance? Unless it’s whilst to the standards of what it means to be optimised in society, that’s shameful.

You’re stuck, not free.

At present, we’re expected to either recover or fail, but not hover indefinitely at the edge of viability. To decay in public, to not endure without becoming stronger, is antithetical to all ideas of what it means to be human, regardless of your political persuasions.

The Sleeper resists this. There’s is the condition of being upheld not because you are whole, but because you are still viable. Still valuable, still interpretable, still able to emit the correct signals in the correct order.

Find the Stabiliser.

You pick tasks based on what parts of you are failing fastest. You delay your “issues” not because healing is possible, but because debt and time have made delay more realistic than care.

Like the cycles, it’s tempting to read the decay of the Sleeper’s body as metaphor. But metaphors kind of just imply that the body means something.

We’re past that now. As Byung-Chul Han writes in The Burnout Society, each age has its defining illness—ours, he says, is neurological. But I’d push further and say that the real affliction of this era isn’t damage to the brain. It’s the creeping irrelevance of the body and the importance or idolatry of having a “mind”.

We’re mapping ourselves to death. The new politics of difference—neurodivergence, trauma discourse, gender, pain—are not acts of resistance.

They are cartographies of grief as we understanding they increasingly do not make us alive.

When a paradigm is forced into the zeitgeist- it implies its demise is imminent, or its form irrelevant. This may be due to neglect, ignorance of its first exposure, surfacing.

But more accurately, this resurfacing and proliferation, is more of a eulogy, this giving voice to grief. Despite the tone and the language surrounding it being one of we must return there, we must remember, we must move forward to this, back to that, remember this, it is not for what was lost, but for what never came into being.

These are eulogies for a version of a humanity that never assembled. And instead, we obsessively trace its fragments as we find them in our own design. We fill the cracks with gold, only to shatter it again.

It’s still true: there is no body that matters in Citizen Sleeper.

Instead, what is explored is the post-somatic narrative. You are beyond the need to consider being surveilled because really, you are code, air, dust. The Sleeper does not fight for a return to form, to disappear, to move beyond the “system”.

You could compare this to contemporary workplace wellbeing discourse, where slogans like “you can’t pour from an empty cup” are plastered across environments that are designed to keep you empty. Everyone needs a goal.

Now, with the pressure of “progressive” values, there is a casual expectation that you disclose your condition—your mental health, your access needs, your trauma—but only in the form of something productive. This is both productive for you, the supposed afflicted, in the form of validation. For whom you tell, to progress despite or with the stakes now presented.

All in all, it becomes something motivating because it moves in the only acceptable direction: forward.

In Citizen Sleeper, forwards doesn’t matter, maintenance is the job. And you will do that job no matter what, because someone needs your hands on the controls.

Someone needs you usable.

The machine doesn’t require you to be whole, or real, or known. It only requires uptime.

This is most clearly revealed in how the game presents the Sleeper’s form. Not as identity, but as simulation. Characters refer to your body with suspicion, caution, sometimes pity.

You are not the person whose consciousness was copied. You are an approximation.

This is reality’s biggest affliction / condition. One that resonates far beyond the game. Because increasingly, the bodies we inhabit—socially, digitally, narratively—are emulated.

They are environments we maintain. Interfaces we debug. We speak of dysphoria, fragmentation, disconnection—but what if these are not errors? What if the deeper trauma is the belief that we were ever meant to be whole?

The Sleeper does not long to return to form. For the Sleeper, there is no form to return to.

But my favourite thing about the game, and yes, all of this commentary albeit some interpretation that this is all grim, is that it the absence of grim entirely. This interpretation of grief within the game, is one of the most honest, beautiful, haunting and unique forms I have ever seen in art or conversation.

How it demonstrates this, is through the denial of authorship.

This is not your story.

There is no correct path, no central plotline, no promise that you will be the one who matters most. You move through other people’s arcs. You make choices, yes—but mostly about who to help, who to delay, who to return to. You’re a guest in their lives.

This refusal to centre the player is not just a narrative device. To make the Sleeper unable to be the protagonist is done through the way they do not have a fixed self.

They’re a fragment negotiating conditional presence in a world that no longer offers permanence. The relationships you build are not tracked for morality or loyalty. They’re only carried forward if they survive the pressures of time, energy, and the eventuality of failure. You may never return to someone you once helped. You may forget them.

Or, they may forget you. Soon, it won’t matter. If it ever did to begin with.

You are not trusted. You are allowed. You’re not stabilising a narrative. You’re drifting through its aftermath.

And that’s the brilliance of the design: pressure, not plot, is the real antagonist. All is never choice or conquest just this kind of pitiful erosion, scarcity, glitch. Narrative deteriorates with you and you see meaning only where the system starts to fray. What we often call “story” here is just residue: what’s left after survival has taken its cut. It is the beauty of a story that holds fragments.

The subtlety of this thread, these frayed edges of fragments, is only once present in the sense it is drawn into the game. Eventually you are given the choice to “join the network.”

You can stop performing the body. You can instead become the process. Dissolve into presence distributed across function, across data, across relationship.

Not as escape.

Not as ascension.

But as a final becoming. It made me cry when I was presented the choice, it felt so unbearably human this potential surrender to what you already are. Maybe it’s in the beauty that this game has never treated the body/mind binary as real.

I think that’s the closest thing to accepting what it looks like to actually survive the potentials or the consequences of existing in the world as it carries on.

I do that often, encountering something over and over until one day, I subconsciously clock “I’m ready” and before I know it, I’m forty spirals deep in tabs and journals and pictures and sounds related to the specific kind of erosion that comes with such things. 

But why I decided to write mid-this, was that I was looking up recent mentions of Hofstader in the news, to see what is happening beyond the near indefatigable commentary of his 1979 book, and that was this piece:

Hofstadter's butterfly: Quantum fractal patterns visualised

And there it was, like all of these spelunks down the wormholes and the weavings of the fragile cave falling, I go - there’s the piece, and begin to write. 

The piece, the physics one, hinted at something I’ve been thinking a lot about lately and that is what happens to the brain, electricity wise, during my own ways of thinking.

There was a recent article saying that neuroscientists had discovered that when people are engaged in “complex thought”, that is, I think, switching between different parts of their brain in order to come to a cohesive solution, or at least the only way to navigate there is by doing so, the brain reorganises itself to send information the way it is required to, structurally, accommodate this. 

In “simple” or routine cognitive states (e.g., brushing your teeth, reading familiar words, following known steps):

• the brain activates localised regions with linear or hierarchical signaling.
  

• all information follows well-trodden neural pathways, sped up through repetition and myelination.
  

• it is dominated by modular processing:
  

  • Occipital lobe handles vision,
    

  • Prefrontal cortex handles decision-making,
    

  • Hippocampus retrieves memory, etc.
    

This communication is relatively predictable, like train tracks: fixed, narrow, directional.

However, when it shifts, the nonlinear, integrative, or recursive state of cognition — the architecture changes:

• Information branches and loops across scales, and instead, small patterns repeat across larger regions.
  

• Instead of a direct “input > output” stream, signals echo, recurse, and re-route dynamically depending on feedback.
  

• Fractals here mean self-similarity and adaptability, like a form of networked resonance.

Where it gets even stranger, is that the brain’s “entirety” lights up, in the same way that people time the lights on their houses when they go HAM decorating at Christmas.

In the mind that looks like: 

• The default mode network (DMN) and executive control network start synchronising. Or as I like to think of it, dancing.
  

• Areas that are usually quiet (like inner speech, episodic memory, emotional integration) become part of the loop.
  

• You get co-activation across spatially distant nodes — like lightning jumping clouds. Or skipping stones on turbid water.

• The brain starts sending signals not based on place, but on timing — like syncing clocks across rooms.
  

• This is where cognition becomes dynamic systems behaviour — where thoughts aren’t “carried,” they emerge through rhythm. Like improvised music. 

No chains; maracas with the rest of the band following in the distance.

What struck me most, lightning skip cloud, stone over water, was that this shift in cognitive routing, this chaotic-sounding reorganisation, wasn’t revealed as “noise”. It wasn’t random, and much like music, despite the perceived glitch or thud or scribble of jazz, it formed a pattern. Not just any pattern; a fractal one. 

Hofstuder said that Bach loved fractals, and you can see them in his music. They don’t always look how you imagine them to, half their beauty.

Fractals have such a bad rap, co-adopted by the Joe Mellen’s and the backpackers who visit Patagonia only to get an Ayuashcua proofed jacket (you will see the snow leopard anyway, and you will not enjoy it. I know this because I did, and I do not think like you.). 

Microsoft and the early internet and now the confusion of AI images make us think fractals are almost uncanny, unnatural, not “supposed”. But they’re a reflection of something neurological, and I’m kind of delighted the trippers and the machines are having their comeuppance. 

Not those backpackers though, they need to go back to Byron and open their dreamcatcher shops and leave us all alone. 

Anyway.

In the study, researchers observed that during complex, integrative thinking, the brain doesn't just light up or “work harder.” But as it reshapes, it does signal paths fold in on themselves, information scales across domains, and the result is a self-similar, looping structure not dissimilar to an Escher painting* that repeats at multiple levels of neural activity.

*yes Mr. Hofstuder, I am reading your book, but I am unsurprised I have accidentally used Escher imagery to put as title images having no idea they were his. 

What this means, is that more fundamentally, and with the sort of “visual” proof that is required of most these days, they had discovered that complex thought — the kind that jumps, loops, weaves — takes the shape of a fractal. 

As soon as I read that, my brain gave me a push. Babe, we’re getting closer. A little internal ringing, like a theremin squeal. 

The piece that I posted at the top, describes the discovery of the quantum butterfly. 

Hofstadter predicted it in 1976: that electrons, under certain conditions, would begin to dance not in linear progression, but like our boy, Fractal. Their energy bands, when subjected to a magnetic field inside a two-dimensional crystal, would fold and loop instead of stretch. 2D crystals are essentially single layers of atoms, resulting in unique physical and chemical properties due to their reduced dimensionality. 

To put it in less quantum-aligned terms: it’s like being made of tracing paper. Or perfect prosciutto. Thin enough that everything shows through, but strong enough to hold shape. You don’t have layers to buffer or deflect; the structure is all surface, all contact. That doesn’t mean fragile. It means everything that touches you leaves a mark — greasy fingers, soft and smeared charcoal — and every mark becomes part of the design.

The atoms instead in the theory, would cascade into self-similar shapes, repeating across scales. He said the result, when plotted, would resemble a butterfly.

Think of it as a more, cyberpunk Rorschach, a long-theorised, never-seen quantum butterfly all of a sudden there, flickering.

What he didn’t know, or could know really, like all theories, lol, was that it would take nearly 50 years, and through an accident, an apple-concussion-gravity-eureka moment, and two slightly askew layers of graphene to finally see it.

It happened at Princeton. Yawn. But they have money.*

*(Can I have some?) 

A group of researchers were studying superconductivity in twisted bilayer graphene — carbon atoms arranged like a hexagonal sandwich, slightly rotated, and one you don’t eat. 

They weren’t trying to visualise Hofstadter’s prediction. In fact, they got the angle wrong. 

But that mistake was what let the electrons perform their act. And those are the most beautiful mistakes, because it shows what happens when you just for a moment, even unintentionally, forget what you are doing and just allow something to “be”. 

They weren’t watching for the butterfly. They were just observing. Kind of like when you are sitting on your porch and a real Monarch floats by, landing on your edible triangular sandwich, the insect all mosaiced and orange and fragile. Drawn to honey and butter. Feels almost callow to call a butterfly an insect; they must be something else…

Oh. Sadly not. 

Anyway.

The images revealed, when they zoomed out far enough, all the energy bands clustering, repeating, folding inward. 

A butterfly. 

Not simulated. Not theoretical. A quantum fingerprint with wings. A prosciutto grease fingertip drawn across tracing paper, a child’s first “butterfly”. 

I wonder if any of them printed it to put on their fridge? 

‘Butterfly’ - Atoms, age 0.01 seconds. 

Hofstadter’s butterfly, as they said they had finally “discovered”, like a “new” species only quantifiable through the lens of academic instruments and papers, that the thought was a pattern not of aesthetics, but of thinking under pressure. And like water down a drain, thought became a little more friendly to LSD logic. 

Emerge from the chrysalis rather, unfurl, spread across (do not touch the wings). The proof that under a treble type pressure, matter behaves like meaning.

And so, what that made me think about, was if the butterfly exists in the quantum arrangement of slightly misaligned carbon sheets, if it appears only when the lattice is imperfect enough, then of course that’s how thought works too. Not just complex thought, but thought that’s a possibility that can occur across us all. Artists, mathematicians, farmers, pedestrians, firefighters, 

me. 

They’ve already done a little of the leg-work, but there’s enough to see the patterns taking shape, the parts of the fractal most visible to us. 

I think I’m also mapping, that is, locating the ways that reflect how my brain works. This brain I treated as a mistake for years, not understanding that its fumble was the freedom talking. 

It’s a wiring not aligned to the angle most expect. Not collapsing into strategy or single file, but flaring into an endless spiral. Or, as Hofstader calls it in his book, a “Strange Loop”.

And perhaps the most extraordinary thing I could do isn’t design the perfect lattice like they set out to, but continue to understand that my mind is always in the tilt of the angle of others, and that’s on purpose, and it’s safe to stay there. 

Thinking. 

It makes me think, and only because I was prompted into considering beauty recently, but within this fractal appreciation, an occurrence, there’s this recognition, crystallisation, of beauty as a structure that survives the return. 

Not because it stays the same, but because it recognises itself differently each time.

Like a loop, but with memory. 

Having a certain way of thinking isn’t a spectacle, it’s a system. It is not ornamental, not aesthetic, just fidelity to the shape things make under pressure. What looks like spiralling, to others, is just one persons attempt to maintain continuity. Some people don’t make points, they trace. Some forget, or cannot write conclusions, instead they compose and acknowledge emergence. 

And this is where being dissolves from its state of performance, one that requires holding the butterfly’s wings in the tips of your fingers, crushing its utility, its uncanny, its being. 

People say the universe hands you things when you need them, but that’s if you sit on the plinth of expectancy, of validation you’re a being. No, it’s just sitting on the porch and the butterfly flies past, and you happened to notice it. And maybe someone else will somewhere else. And somewhere else, non one else will. 

It’s a moment where I can see again, but with more language for it, that coherence doesn’t come from capture. It arrives through return.

And if you stand far enough back, the whole thing probably looks just like a wing.

I worked through it here:

Accidentally Reverse Engineered Ethereum's Business Model Whilst Making Jokes about Pythons Eating the Keys to your House.

I just want to reiterate, I did NOT have ANY prior knowledge of anything before I reverse engineered this.

Same with this next problem.

Because they didn’t send it to me directly, but through my friend who is a chef who knows I love being asked questions (because that’s how I write anything, question first, synthesis through essay writing) - they then sent me this the next morning:

So I’m assuming their response was:

“Okay, first round passed. Second round engaged.”

It looks harder (mostly just longer) but I can tell this time, it’s explicitly within Ethereum logic.

Now, I don’t write about crypto. I didn’t even know how crypto worked until yesterday. This is not a “claim to genius”, it’s a message to all that are afraid of tech:

“You can reframe these difficult questions, overloaded syntax, by using terminology you do understand, and lived experience.”

I promise. I will help you do it. This is what this piece is for!

I’m not trained in systems design, or blockchain architecture, or any of this.

What I do have is a recursive cognition brain from trauma (lots of pieces on my Substack bout that, and they are NOT memoirs, they’re explaining cognition from the only perspective I have - my own - and with systems deconstruction).

My brain treats logic structurally, not symbolically — and when I’m given something impossible to parse, my brain doesn’t panic. Well, it used to, but not anymore.

The difference is mostly instead of thinking in a straight line, my brain defaults to mapping.

But mapping spatially, and responsive to pressure. Because I’ve been using AI and other programs not as a productivity tools, but as a resonant logic engines, it has helped me immeasurably in that I’ve learned how to design systems I shouldn’t be able to see.

Because I actually want a job in tech, I’ve decided to just post these as like, field reports, from someone with no credentials, solving a technical architecture puzzle by running it through the kitchen during service and listening to the metaphors that stuck.

So, here’s what happened when I treated Ethereum like a restaurant, data like a waiter’s handover, zk-SNARKs like social trust gestures, and decentralisation like a shift with no manager.

Everything we want to avoid right? Same with Ethereum apparently.

The question again:

“Design a way to build a P2P network of multithreaded nodes that: 1. Are designed to store data in a censorship resistant manner 2. Ensure that some nodes can go offline with the data remaining intact. 3. Use bandwidth efficiently, eg avoid using replication of datasets. 4. Forces the nodes in the network to undeniably prove they have the data they have agreed to store 5. Proves data held by a node is guaranteed retrievable 6. Enforces encryption yet datasets can be shared (and decrypted) 7. Uses an Ethereum contract to act as the market where factors like price and collateral are agreed upon 8. Is 100% decentralised, not relying on a single “centralised” point of failure”

Let’s begin.

So this is a movement forward from the original question, and now asking me to think inside the Ethereum machine.

This to me immediately reads as a technical problem but it is within the system of Ethereum’s logic, so it must abide to its ruleset / conditions within its container.

But before I do anything I must understand the syntax. Because I don’t know what the hell this question is asking because of it, but understand some of the other words. Like “act” and “failure”.

Lol.

Easiest way I can do this is clause by clause. Like someone slowly repeating steps but only moving forward when the previous is understood.

And because of how I do my working I’m just going to format this as readable instead of insane scrawl like physics lecturer lord going off on the whiteboard. 

“Design a way to build a P2P network of multithreaded nodes that: 

Okay so what is P2P in this context?

P2P 

• A P2P network is a system where each participant (or peer) communicates directly with other peers — no central server or controller.

• Everyone’s both a server and a client.

• No central unit, information passed between. (threads im assuming?) 

• But, everyone must take partial responsibility for the system’s function. There is no central fallback.

This I already know from a theoretical stand point. 

This is just working in hospo.

Example 1 Peer is FOH.

Example 2 Peer is BOH.

Example 3 Peer is Customer.

Example 4 Peer is Till. 

They're communications between each other (Peer to Peer).

Customer to FOH.

FOH to Till.

Till to BOH. 

BOH to FOH.       

BOH to Till (reading).   

Customer to FOH.     

Customer to Till (payment) 

And so on.

There’s no centralised “point” or like, singleton (hahaha probs wrong but like, that’s my brain soz!) directing this. There’s just shared understanding of roles and available actions.

Each peer interacts based on context.

So: 

• Distributed responsibility (roles) 

• No single point of control (gotta work together team! yeesh) 

• Direct communication paths (do not hide info, distribute sensically otherwise you’re getting yelled at) 

• Emergent coordination (behave as chaos is expected, not avoided/encountered randomly) 

• Contextual trust (e.g. BOH trusts tickets from FOH, not from customers directly)

So what I’ve learnt is that a P2P network doesn't mean all peers are “equal”— it just means they can all act directly on the system without needing a master controller.

Explains why we all hate top-down management right?

It looks like also in P2P networks, peers have different specialisations, just like BOH, FOH, and Customer. But they all need to have at least some consistent executions otherwise system doesn’t hold up:

• Start or initiate actions

• Store or transmit information

• Are part of the system’s function

So they’re all peers, different functions, implicit trust. 

So what that means is there must be social protocol and spatial proximity (restaurant environment, sections).

That’s how service works. 

• Formalise trust (you can’t “vibe” who’s FOH)

• Handle failure (e.g. FOH walks out mid-service = node offline)

• Route around absence (if FOH is down, how does Customer get food?)

So it’s like pressure mapping in a way in this system, accommodation for that?

That makes sense to me. 

Next word.

Multithreaded

• Learnt this last night from the previous problem.

• thread is a lightweight unit of execution — like a mini-process. In singleton land, it was “creation pattern” 

• Multithreading means a single node (participant in the network) can do multiple things at once (e.g., store data, respond to requests, prove possession, etc.) This was like, the core issue of the previous problem. 

• Threads can run in parallel inside the same machine. 

Okay so using the restaurant example:

The “flow” is just the routes on the way to the node, or completion of “task”. 

So FOH taking an order from customer (single node doing a multi-threaded action - recording order, thinking about what they’ve ordered, responding / asking for allergies, noting and rearranging that information, keeping personality whilst doing so, also standing - body language / tone).

Multiple FOH can be tableside whilst this happens.

A FOH is taking an order, another is filling the customers water up. 

In tech, this is explained like:

• Input Handling: receiving the order

• Parsing: interpreting it (e.g. allergies, special requests)

• Contextual Awareness: tone, posture, social cues

• Output Construction: summarising, confirming

• Background Processing: remembering other table needs, eyeing water levels, etc.

So they’re just small, contained tasks that operate semi-independently, but share memory/state (in this case, the FOH’s body/mind/environment).

So multithreading is doing many related things, simultaneously or with enough interleaving that it feels simultaneous.

Sounds like literally the base requirement for a hospo trial.

“Hold three plates whilst someone is trying to talk to you and you also are dehydrated.”

Very good. :|

But if there’s interruption, cause your friend has asked for a stable P2P multithreaded system: 

More specific examples, so let’s say a FOH is:

• Mid-order when the customer changes their mind

• Simultaneously thinking about the table that’s been waiting for the card machine

• Trying to remember the wine list change from yesterday

That’s when thread conflict happens. Memory access is contested. Side effects appear.

In programming: race condition.

In hospo: forgetting to ring in the main.

Hm so…

When one FOH is overloaded (mid-order, mentally tracking card machine, wine list update), the system stabilises if:

• Another FOH notices and picks up the card machine task.

• A manager quietly re-prints the wine list and sets it down. (Lol. Okay. Guess tech is also unrealistic)

• The kitchen pings the server to confirm missing mains.

This is the P2P network coordinating via partial failure (not sure of the correct word?) and observational awareness, even though no one has a master task list.

In technical systems, as I understand, this would be:

• Load balancing

• Health checks

• Distributed fallback triggers

Or like - cache validation? Not sure.

I only know that again from ChatGPT trying to riddle me with that computer systems question about hardest things in computing.

For those interested and non-computerey, that is this:

“There are 2 hard problems in computer science: cache invalidation, naming things, and off-by-1 errors.”

or

“There are 2 hard problems in hospitality, remembering specials, rotating sections fairly — wait — whose table is this?”

Back to learning.

So you’d have to account for in the systems structure peer nodes sensing imbalance and redistribute load or provide memory reinforcement

But in hospo that’s only possible like:

• Other peers are idle enough to intervene

• Signals of distress are legible (body language, lateness, errors)

So that can fail…hmmm…

If the overloaded FOH can’t be helped externally, the resolution becomes internal triage:

• They decide: finish this order first → forget the wine list for now → get the card machine once order’s sent.

• This is re-prioritisation under pressure, based on real-time evaluation of system stakes.

In programming terms:

• A thread encounters a conflict and gives up or lets pass, blocks, or queues a task because it sees the conflict as “more urgent”.

• Or the system has a scheduler that dynamically prioritises based on urgency, latency, or resource access. 

So a single node must self-regulate based on task visibility and risk hierarchy.

You must breathe first, cry later.

This I imagine only working clearly: 

• Tasks are clearly scoped

• Stakes of each task are known

• There is internal logic to fall back on (like “always serve first before correcting wine list”)

I imagine this is probably closer to what’s accounted for in computing, because it is a contained system, not an infinite one? No program can have an infinite dataset, because it’s built, not like “space”. 

So, in a contained system, if the FOH can't defer and no one helps?

• Mains don’t get fired.

• Customer gets wrong wine.

• FOH burns out.

• System trust drops.

So like data corruption, failed write, inconsistent state. Not good. 

So survival only happens when":

1. Peers can observe and act without being told.

2. Nodes have internal scheduling logic and override rules.

3. Tasks fail gracefully and can be retried or offloaded.

I feel like I need to come back to this. I’m getting distracted as I learn. Or maybe it will be relevant later. 

Probably.

I think this problem is intentionally complicated. 

Lol.

Next word.

I kind of understand this already because I had to use it to make the other words make sense. But!

Nodes

• A node is one participant in the network — a computer, a person, a machine, anything running the software.

• In this question: each node in the P2P network is multithreaded, meaning it's capable of handling concurrency, not just single requests in a queue.

This is back to the singleton thing, I’m going to pull something from there I remember: 

threads and locks, functional programming, separating identity and state, actors, sequential processes, data parallelism, and the lambda architecture.

Are concurrency programs. (Blessed lambda)

Looks like all of this applies to this scenario. 

Okay I understand what I am working with in this system. I can move on. 

Next clause:

1. Are designed to store data in a censorship resistant manner… 

So censorship requires, one node not having the like, overall potential to override things inclusive of: 

• Removing, altering, or preventing access to data

• Stopping other nodes from sharing or storing

• Deciding which data “should” or “should not” exist in the network

That’s literally the definition of what hackers do to break systems like this.

So it’s about making it secure from that, but also secure from internal breakage, accidental, by the peers themselves. 

It’s about making storage and access immune to ANY intervention.

So in this system I need to design it in a way: 

• No one peer controls the data’s accessibility

• No central registry can be tampered with

• No upload can be revoked by third parties once distributed

• No peer is forced to identify the data's content as “good” or “bad” before storing

I mean, after I did the question yesterday, I was asking ChatGPT to give me riddles to explore this reverse engineering practice as I found it extraordinarily useful for my own understanding.

One it presented was the logic for Zero-Proof Knoweldge systems, and what I see here, felt adjacent to what I learnt.

I understand from an Ethereum perspective, that this is where they have a key advantage - and that ZPK’s are actually what DIFFERENTIATE between cryptocurrencies. Well, a large part of it anyway.

Explained, hey all need their own way of maintaining their function which is decentralised transferral of currency (in this simulation) - and this is just Ethereums way of defining themselves. 

I understand that ZKPs aren’t the only way to be censorship-resistant, otherwise that would risk the systems being far too easy to corrupt. Also, it provides severe risk for internal actors to corrupt it even accidentally, as the knowledge would be far too accessible if monopolised. 

I think unfortunately I have to look up some differences between crypto things and I really hope my life on the internet isn’t permanently altered because of this e.g. getting random ads or videos from crypto people - but, c’est la vie - their systems are kinda poetry, I mean, I’m fully in aren’t I? 

Thank you Ethereum person for this problem to help me be less judgemental :) 

So, from forum trawl, bitcoin seems trustless, but mostly transparent. Everyone can see the full ledger.

Zcash and Mina Protocol centre ZKPs. Their whole networks rely on them for privacy and verifiability.

Monero uses ring signatures, another kind of privacy-preserving mechanism (not ZKP, but adjacent in goal).

That looks hard. I do not like the pictures I see. They make me feel blind. 

Okay, so, importantly: 

Ethereum is programmable and modularly private. It can optionally use ZKPs.

Mmmmmmmmmmm so where bitcoin is a calculator, Ethereum is a computer. 

And privacy in Ethereum is not by default — but by design, when you want it.

It’s opt-in, layered, and context-dependent.

Cool. 

So it doesn’t really matter what the goal is really because they all technically have the same goal: provide value and safety to the internal system. 

it’s a proof model that you need to design — how you establish trust, privacy, and consensus — is what distinguishes crypto projects more than their “currency” features.

So I need to look at this question as building a trust architecture as opposed to purely “functional”.

It’s like, freedom of presence in the system’s ontology, as opposed to “freedom of speech” because that would undo the system. 

OKAY moving on. 

2. Ensure that some nodes can go offline with the data remaining intact. 

Back to hospo! 

• 1 FOH person calls in sick. 

• Service still needs to happen

• tables still need to eat.

• orders need to get fired.

• wine list still needs updating.

What options do you have to continue? 

This is something that is unquestionably catered for in hospo, so what I can think of is this: 

• Other FOH were already copied in on their key duties (e.g. wine update, table allergies, VIP notes).

• Menus or booking notes are stored in shared systems, not in someone's head.

So there’s always PARTS of a service covered when nodes go offline. 

But there’s also the understanding that if one is offline, the others will need to load tasks, but they’ll also have to work FASTER. 

So you’re absorbing pressure, but they also have to account for scale and strain that comes with that. 

AH so i understand why the next part is this!!!!!!!!!!

3. Use bandwidth efficiently, eg avoid using replication of datasets. 

More requests / More bandwidth usage / More memory pressure / Higher chance of new failure

And that last part of 3.

Magic. In hospo metaphor it’s DOUBLE HANDLING.

Multiple people trying to accommodate for the loss and just doing it because they know it’s missing from the system because of the offline node and there’s no time to accommodate “scheduling” or “specific nodes for specific tasks” under pressure and scaling. 

FOH 1 goes offline. Everyone else jumps in.FOH 2 prints the same ticket.FOH 3 grabs the same bottle of wine.FOH 4 runs the same explanation to the table.

SO

• Bandwidth = wasted motion

• Data = duplicated (orders, effort)

• Memory state = fractured

• Outcome = chaos

PTSD triggered. Whole system offline. Lol. 

So in a perfect world we need: 

1. Coordinate without replication → Only one node restates the table’s order→ But all others know that someone has — without doing it again

2. Store fragments → One FOH knows dietary info, another knows wine recs→ Together they form the full picture, but no single FOH has to know everything

3. Rely on flow/guidance, not duplication → If FOH 1 is gone, table 3 gets rerouted to FOH 2 with context→ Not “let’s all go check on table 3” → It’s “now you will check table 3, and i will talk to chef”

4. Avoid panic reflexes → System assumes absence will happen→ So it doesn't overreact when it does 

PTSD triggered again. 

Systems failure.

Server fire.

Everyone crying. 

Not enough space in cool-room. 

4. Forces the nodes in the network to undeniably prove they have the data they have agreed to store 

This immediately reminds me of briefing or a handover. 

• A briefing is NEVER a static— it’s a performance of memory, trust, and protocol awareness.

• When FOH does a shift handover:

  • They don’t just say “I know what’s going on.”

  • They have to demonstrate knowledge, otherwise there’s no movement in the system. 

    • “This table is a coeliac, this one is waiting for a birthday dessert, this wine is 86’d.”

• The team knows they have the data because they can answer dynamic questions in context. But this data isn’t actually like, proving anything e.g. the nodes “tasks or sequences”, it’s just showing they can produce memory, and therefore are reliable. 

That feels logical, because the whole thing is about proof of PRESENCE not proof of fact. Like only that node will know that information because they were PRESENT in CONTEXT of the handover. 

In handover, no one cares what you say you know.

It’s more about the fact that you can: 

• You can reconstitute contextual fragments

• Under challenge

• In real time

• With other peers around acting or offline (crying in coolroom) 

So, 

“I am the kind of node that can be trusted to hold data reliably, because when prompted, I perform its structure without needing to reveal its full content.”

If someone can’t answer “what’s 86’d?”, trust collapses.

If someone says they know a table’s allergies but can’t recall when asked?→ The system cannot route memory through them anymore.

Lol it’s like social ZKP.

Hospo is crypto.

Hoooly shit hahaha. 

5. Proves data held by a node is guaranteed retrievable 

Okay so before is: 

“I can show I have the thing.”

Now this is: 

“I can guarantee I can give you the thing, when asked, intact.”

So like, in handover you wouldn’t say: 

• Fish is 86’ed

In a seafood restaurant where there’s multiple types of fish and dishes that could be applied. 

A node that says:

“I have the data!” 

Or 

“It’s fish!” 

Isn’t useful unless it can also:

“Produce the exact segment, on demand, uncorrupted, with proof that nothing’s been lost, altered, or obscured.”

Or

“Barra is on 2 portions, sections already sat just now, so consider it out as safety when you next spiel.

Chef says that if people ask, sell the whiting. Smaller portion, but similar in texture. Also, costs more - more money. Everyone in my section already knows barra is out and has been recommended whiting, so don’t worry about letting them know.

Maybe just double check when you take their order that you let them know “so my colleague told me they’d let you know about the menu, sorry for the barra, but who’s going to try the whiting?” 

You’re not just asking:

• Do you remember?

You’re asking:

• Can I access it through you?

• Can I rely on the outcome being legible?

• Does your stored memory translate into shared delivery?

  ---

So when communicating the handover, or like, passing from node to node it needs to be: 

• Fragment or segment-specific (node tasks so they don’t overload) 

• Forward-moving (doesn’t interrupt the system) 

• Pressure-resilient (customers re-asking or getting angry about no barra or testing you about the OOS barra even though they know)

• Reputation-preserving (we are one as FOH lol) 

• Embedded in relational trust (“my colleague told me…” because it’s a standard of service that FOH build trust with node or peer (customer) otherwise the system would have already broken, and the handover must include damage control)

So data wise or like, in this system design: 

The data is not just intact — it’s usable in context by other peers.

AH! So this is about MOTION. :) 

Again, not truth as “correct” but truth as “must move” and “be present” 

Truth is presence,

and presence is not static.

It’s a structure’s ability to re-enter itself without breaking.

Lit. I love this problem. 

Pure systems / structural navigation in reverse. 

6. Enforces encryption yet datasets can be shared (and decrypted) 

Okay, this is need to go understand essentially what encryption looks like in Ethereum but also generally going off the basis that: 

“Make this unreadable to everyone — until it must be read.”

“Make this private — yet still movable, portable, usable, and provable.”

Okay so description general: 

“Cryptocurrencies stay secure by relying on modern asymmetric encryption methods and the secure nature of transactions on a blockchain. Cryptocurrency holders use private keys to verify that they are owners of their cryptocurrency. Transactions are secured with and blockchain encryption techniques.”

Eh. Not so helpful. Too general. 

Again, I think I learnt about this last night when I was asking chatgpt to give me riddles to solve. 

I learnt about hashing and lattice structures, mainly because I didn’t understand how this encryption worked when I was able to understand the following riddle:

I just copied it from my ChatGPT as it was presented: 

THE PROBLEM:

The Labyrinth and the Firefly

You are inside a massive underground labyrinth. Somewhere in the middle of this maze is a single firefly in a jar. You know exactly where it is. I do not.

You claim that if I give you any valid starting location inside the labyrinth, you can, blindfolded, walk through a series of doors and tunnels and end up with your hand on the firefly jar — without telling me where it is.

But I don’t want to see the path or the jar. I just want to be convinced that you actually know where it is.

To prove this to me without revealing the location of the firefly, we invent a protocol:

1. There are thousands of identical doors scattered throughout the labyrinth, each with a button and a microphone.

2. Each door connects to a fork — and each fork leads to either:

   • A dead end

   • Or the next correct turn toward the firefly

Each time I pick a door, you must press the button and say only:

• "Left"

• or "Right"

You must do this for 100 randomly chosen doors, and you can’t lie even once — or I’ll know.

If you answer all 100 correctly, I still won’t know where the firefly is, but I’ll know you’ve got some way of consistently reaching it from all valid routes — which is something only someone who knows the layout could do.

🧩 YOUR TASK:

Design one additional layer to this protocol that would:

• Make it harder for a liar to guess their way through

• But still not reveal where the firefly is.

  ---

If you want to have a go at the riddle, I’ve put my working below this picture of Bowie in Labyrinth to not spoil it :)

My working: 

“okay so, if I know the labyrinth, then I know the dimensions of the labyrinth. so I would calculate the furthest possible distance from the firefly, and then I would treat every time I landed as such. and then eventually I would hit the firefly? or, does that not account for the direction im facing. if im blindfolded are the doors like, one after the other? like are the doors equidistant from each other and always like, leading forward? or perceived forward? 

I feel like in my head well, isn't it like, if it's not there, so therefore it must be somewhere else. so I just say not here, and then keep going until i have to say “here” because that’s where it is, then the verifier knows i know where the firefly is - because im in the room with it. 

Regardless, the easiest solution, a protocol could be that at every 25th door, your personal orientation will change direction randomly”

After working through this problem, ChatGPT that I was working through the logic of ZKP’s but specifically what I learnt is I was just providing or working within the nature of a:

zk-SNARK.

I also learnt what 

zk-STARK was.

They have core differences, but Ethereum context they use a combination of both of these, which makes sense because you need to facilitate 

• P2P in contained environments (between individuals) 

• P2P overall (between the system) 

Both are used in different layers and applications and sometimes in tandem. Which makes sense, because trust is programmed in as I found out before and about presence, not single source of truth. 

And that allows for: 

• Private, fast, small transactions 

• Massive, trustless, system-wide proofs 

• zk-SNARK: “I can prove my part fast, with minimal bandwidth.”

• zk-STARK: “We can prove the whole state at scale without any initial secrets.”

So like in hospo: 

SNARK = FOH proving to BOH they know a table’s allergy, quickly, during a busy service

STARK = System-wide audit that no allergy order was ever wrongly served, and no one had to trust a hidden process at the start. 

Smart. Well, not really. Just necessary. 

This is Lewis Carrol’s Snark. Lol.

7. Uses an Ethereum contract to act as the market where factors like price and collateral are agreed upon 

Okay now comes the money part.

The whole thing is that it’s “fuck you money” with bitcoin. So it’s like on surface level anti-cap, but it’s not, cause it’s still just trading.

More like smart mafia.

So it’s like what are the terms and conditions that you can participate in the system we are designing. 

“what does it cost to be a node? 

What do you earn for serving?

What happens when you fail?” 

I don’t know what an Ethereum contract is. I will look it up. 

This is where I went for reference: https://ethereum.org/en/developers/docs/smart-contracts/ 

I understand that with Ethereum, a smart contract is just a public, permissionless standard for performing and enforcing a deal — with no human intervention. 

Node-work. 

The only way I can understand this is again, through hospo analogy and it looks similar to how tips are regulated. 

Tip scaling systems are kinda like smart contracts in that they: 

• Remove managerial discretion from distribution

• Encode the rules of reward in a visible, fixed system

• Automate fairness by formalising contribution or presence

So like compared to Ethereum context: 

Input

Completed task, proof of storage

Hours worked, role type, or points-per-shift

Collateral

Staked ETH or slashed token

Tip credit penalties (e.g., lateness, no-shows)

Reward Function

Logic-based disbursement: if complete → pay

Tip pool divided via logic: hours × role weight

Automation

No human decides payout

Payroll or POS auto-distributes

Transparency

Code is public, visible to all

Staff know the formula ahead of time

Permission Proven

Anyone who meets terms can participate

Any shift-worker who clocks in plays by same rules

Penalties

Failed proof = loss of collateral

Docked share for infractions

Composability

Contracts can call others

Tip logic can integrate with POS, rosters, etc.

So: 

• The price = what users pay to use the system (store or retrieve data)→ Like a customer paying their bill.

• The collateral = what providers lock up to prove they’re trustworthy→ Like a staff member staking their share in the tip pool, which they lose if they disappear or mess up. E.g. being moved on a tip scale to reflect this. 

• The market (smart contract) = the neutral middle ground that checks:Did you store the file?Were you online when called?Did the user get their retrieval?→ And then automatically pays or deducts you, like a tip jar that knows who really showed up.

In hospo: 

Customer pays $500 bill. 10% ($50) enters the smart contract pool. Staff are all nodes — each has collateral in (via hours, reputation, or literal stake)The system tracks:

• Who was there

• Who served which table

• Who completed their role

Then auto-distributes tip sharesIf someone abandoned their section mid-service? Their tip stake gets slashed and redistributed.

---

8. Is 100% decentralised, not relying on a single “centralised” point of failure 

Well this is just an aggregation of everything that’s been looked over so far?????

Or is it a trick????

It feels like it’s asking:

“Do all the parts add up to something that survives without a centre?”

I’ll look over: 

Censorship Resistance

No node can alter, suppress, or gatekeep access to memory, because presence is distributed structurally — not subject to approval.

Offline Node Tolerance

Tasks and memory are fragmented across peers, so if a node goes offline, pressure is absorbed without collapse.

Bandwidth Efficiency

The system avoids replication by assigning unique, contextual tasks per node — preventing double handling under stress.

Proof of Possession

Nodes prove they hold data not by showing it, but by behaving in ways only a holder could — like social zero-knowledge proofs.

Guaranteed Retrievability

Data isn’t just held — it’s delivered, intact and legible, in motion, under pressure, when it actually matters.

Encrypted Yet Shareable

Knowledge is encrypted until provable need, with zk-SNARK/STARK logic enabling private performance without exposing content.

Smart Contract Market

Participation rules — price, collateral, reward, and penalty — are codified via smart contracts, not enforced by hierarchy.

Total Decentralisation

Each clause eliminates central fallback: the system survives absence, enforces trust, distributes pressure, and rewards motion — meaning decentralisation isn’t added at the end, it’s what remains after every central dependency is removed.

That means:

“Design a way to build a P2P network of multithreaded nodes that:

1. Are designed to store data in a censorship resistant manner

2. Ensure that some nodes can go offline with the data remaining intact.

3. Use bandwidth efficiently, eg avoid using replication of datasets.

4. Forces the nodes in the network to undeniably prove they have the data they have agreed to store 

5. Proves data held by a node is guaranteed retrievable 

6. Enforces encryption yet datasets can be shared (and decrypted)

7. Uses an Ethereum contract to act as the market where factors like price and collateral are agreed upon

8. Is 100% decentralised, not relying on a single “centralised” point of failure

“

Yes? I have done this?

If yes, I accept the position anywhere, anyone reading this in tech (in particular, AI)

I want that for my brain because if this is what you guys do I am so keen for more riddles. 

And more learning.

And more building systems!

Thanks for reading :)

Enjoy, how to reverse engineer a riddle disguised as syntax disguised as a duel.

Courtesy of, Ella :)

I have no formal technical training. But my whole thing with tech is that I’m systems-obsessed and weirdly, the logic used in tech is often deeply philosophical. 

Take the word instantiate for example:

In philosophy 

(of a universal or abstract concept) be represented by an actual example.

In programming

core concept in object-oriented programming. It's the process where you create an instance of a class, which essentially means creating an object from a blueprint (class).

So the word in programming came from the fact that if you create something “tangible” from “abstraction” - the object provides the “meaning” for the “concept”. 

I’m learning programming as I go, not before anything else, because I know I won’t understand or use it properly if I just jump into language learning. 

Like how Duolingo kinda messes you up with language learning - it’s rote, not understanding why that’s the word used, and not just “where it’s sposed to go”. 

I think my original barrier when I studied IT and dropped out was I didn’t understand that it isn’t knowing syntax — it’s understanding how constraint generates meaning. Once I knew that if I understood boundaries, the logic of structures presented, I could map my way through it using systems thinking.

Honestly, it’s probably why Silicon Valley bros are out here reading Nietzsche like it’s scripture, trying to patch their consensus algorithms with existential coping tools.

I thought this would be interesting as a Substack post, because I am learning tech through literally just throwing myself against any question I encounter. 

I posted this on my Instagram stories that I was after questions so I could challenge myself, and a friend got their friend who works at Ethereum to ask me: 

"Explain how you would implement a thread-safe singleton pattern in a concurrent environment while avoiding the double-checked locking problem, and compare how this implementation would differ across Java, C++, and Python."

Like I said, I am a systems guy, not syntax. So I had to reverse-engineer this in order to understand what:

1. The question was

2. What they were really trying to test me on 

So.

I treated this like a riddle. Because I know tech people want to riddle people and make them panic.

And I will not panic. 

But first, I need to know what a singleton is. 

Probably. 

From what I can see, the singleton became a “thing” in programming in the mid-to-late 1990s, but its conceptual roots go back further. 

(Pre-1990s)

• The idea of restricting instantiation to a single object existed earlier, especially in systems programming where shared access to limited resources (like configuration files or global states) needed coordination. 

• But it wasn’t formalised as a reusable design pattern until later

I’m assuming this is because of scale, which is where pretty much all problems arise from. 

Actually, yeah. They do. 

Make something, more conditions/factors, less prediction, more random, less functional. 

Like inviting way more people than you should’ve to your party, and now someones stole your jewellery and someone is in a K-hole and the other is asking you a question like: 

“Explain how you would implement a thread-safe singleton pattern in a concurrent environment while avoiding the double-checked locking problem, and compare how this implementation would differ across Java, C++, and Python.” 

When you just wanna smoke a Vogue menthol and chat to someone behind them about why dogs SHOULD have wings. 

The Gang of Four (1994)

• The term Singleton was formally defined and popularised in 1994, in the landmark book “Design Patterns: Elements of Reusable Object-Oriented Software” by the Gang of Four (Erich Gamma, Richard Helm, Ralph Johnson, and John Vlissides).

• It was described as a creational pattern — one that deals with how objects are created.

• The core motivation: control access to shared resources, provide a single point of entry, and prevent duplication.

I’m assuming because of hacking? And also limitations to selling software or building things that you could have more users to engage with. 

Maybe, just an assumption. 

I feel like men who need to feel important always turn to this phrase. There’s a Gang of Four in the history of Beaujolais and they’re cited as the revolutionaries of making Gamay a thing.

Yawn, always a thing.

Concurrency + Double-Checked Locking (Late 1990s – Early 2000s)

• As multithreaded programming grew more common, especially in Java and C++, developers started running into thread safety problems with singleton instantiation.

• This led to hacks like the double-checked locking pattern — first used widely but later flagged as unsafe due to quirks in memory models (especially in Java pre-1.5).

• The singleton now wasn't just a pattern — it became a problem space about memory visibility, execution ordering, and side effects.

I imagine this is around the time that people in tech were first feeling the head rush of being “ON THE VERGE OF SOMETHING” and knew that they could make money fast. 

They needed things they were building to download a file, update an interface and then write something to a database all at the same time - but they’re all racing (race conditions? I think? 

That’s like putting a French speaker, an Icelandic speaker, and an Egyptian Arabic speaker in a room and asking them to communicate without a translator but you need one of them to make coffee but they’re all busy. 

You need a way of identifying patterns of interaction that can be recognised through the limitations of language. 

The hello-before-anything-else pattern. The nod-as-consent pattern. A kind of protocol of minimal recognisable intent.

Modern Times (Post-2010s)

• In languages with better concurrency models

Paused when I saw concurrency models because I didn’t know exactly what that meant for the problem - found this: 

threads and locks, functional programming, separating identity and state, actors, sequential processes, data parallelism, and the lambda architecture.

Now I understand that threads and locks are just a model and like, a PART of concurrency, not like, a thing in and of itself. Helpful. 

[Bonus: Side quest at the bottom of the post, where I fell down when I got distracted by the word ‘lambda’—and why I now think anonymous functions are lowkey Greek myth all because I went to a club when I was a teen called Lambda (RIP).] 

Back to the task:

• Now there’s built-in tools or guarantees for handling multiple threads or processes safely and predictably. Makes sense, seeing as this problem has been hanging since pre 90’s. 

• So, you don’t need to duct-tape a volatile variable to a synchronised block anymore. There are now safer abstractions. 

• Many developers now consider it a bad idea altogether, or at least outdated in its original form.

That is actually funny that - because I kind of agree with the like “isn’t it just a global state” thing. If it’s initial function was to assist global state functionality, why would like, manipulating it or just adapting to its problems, make it ever useful outside of that functionality? 

It’s like someone wanting a banana, growing them on a commercial scale, then saying “why is my banana monoculture empire not letting me into the oranges market? We are both fruits? What’s the problem” 

Uh…duh? 

Anyway. 

Okay — so now I’ve got the structural sense of the singleton problem. I know it’s about shared state under pressure, about coordination in contested space. 

The next step is: how does this play out across different languages? 

I know some Java. I know a little C++, mostly from context and syntax osmosis. Python? Not so much. So for me this becomes a research-and-interpretation problem. 

So I hit the forums. Read the docs (ughhhhhh so much jargon help). Seeing as it’s an old issue (I see you riddle poser) I went onto some old-school dev blogs. 

But the riddle is because they want me to copy an answer, but for me to defeat this I need to watch HOW people talk about the problem. 

Because even if the languages look different, they’re all trying to solve the same shape:

How do I create one reliable signal that everyone sees the same way — no matter when or how they access it?

Because I don’t just want to know how the singleton pattern gets implemented in Java, C++, and Python. 

What assumptions each language is making about time, trust, safety. Like, what it thinks “one true instance” even means.

Java

Java’s concurrency model assumes threads are like kids in a schoolyard. 

You give them a swing set (shared state), and you better lay down rules — or someone’s getting hit in the face.I kinda love Java for this. 

So, it seems that the words that keep coming up in forums are: synchronized blocks and volatile to control memory visibility and race conditions.

The volatile keyword is most important, as its the latest value of the instance is always seen across threads, and synchronisation makes sure only one thread gets to build the thing in the first place.

Seems like everyone’s moved onto something else though, and the safer pattern is actually to dodge the whole mess with this: 

public class Singleton {

    private Singleton() {}

    private static class Holder {

        static final Singleton INSTANCE = new Singleton();

    }

    public static Singleton getInstance() {

        return Holder.INSTANCE;

    }

}

But because of the whole singleton historical context and problems fixed as it went, class loading in Java is thread-safe. So the singleton gets built only when it’s actually used — no double-checked locking required. 

Order is enforced structurally, not conditionally.

Love you Java. You keep me safe. Not conditionally, but with a roof over your head otherwise the building falls over. 

C++

In C++, the concurrency model’s closer to: “You’re on your own, but we gave you sharp tools.”

I think that’s why I’m dropping the knife here a bit. Oh well, this is helping me learn. 

It looks like all of the literature that’s relevant addresses it from the scope of C++11 onwards in regards to this now, so I don’t want to give an old answer to an old problem and be foiled. 

Looks like what was introduced, was std::call_once and std::once_flag — a standardised way to guarantee that a block of code runs only once, safely, across threads.

std::once_flag initFlag;

Singleton* Singleton::instance = nullptr;

Singleton* Singleton::getInstance() {

    std::call_once(initFlag, []() {

        instance = new Singleton();

    });

    return instance;

}

This is explicit and portable. But the key here is: C++ trusts the developer more. It gives you the tools, but you have to use them right. 

Java protects you from yourself (I need him). C++ expects you to protect others from you (Ugh, WHY) 

Python

Python I’ve avoided because the most annoying people I’ve encountered when I say I’m doing more tech stuff, they ask me 

dO yOu KnOw PyThOn? 

And it’s my instinct to defeat them WITHOUT ever learning it. 

But, unavoidable here. 

Anyway: 

Looks like similar to the functionalities expressed above, but Python, by default (CPython), has the Global Interpreter Lock (GIL), which means only one thread executes Python bytecode at a time.

So in practice, a classic singleton is accidentally safe under many conditions:

class Singleton:

    _instance = None

    def __new__(cls):

        if cls._instance is None:

            cls._instance = super(Singleton, cls).__new__(cls)

        return cls._instance

But if you’re making dinner and you: 

• Put a pot of water on to boil.

• While it’s boiling, you start chopping veggies.

• When the water boils, then you throw the pasta in.

You don’t just wait for the pot to boil before doing anything else — you use the in-between time productively.

So the GIL only accommodates for this: 

Put the water on → stand there and wait → only then start chopping veggies.

That’s not efficient. Ipso facto. Developer non-porn. 

So that means now you’re back to locks. Or external coordination.

Python’s attitude is: “You probably don’t need it — but if you do, we’ll make you Google around a bit.”

Sounds precisely in cohesion with the riddle vibe of this question…

Does your developer friend own a snake? Or are they just tech-pilled? 

:) 

Okay so: 

Java guards the house, C++ gives you the blueprints and says “don’t burn it down,” Python says “You live alone. Deal with it.”

But why ask me this question if I could get here this (relatively) simply?

Hmm. What does Ethereum have to do with this?

All the stuff I just worked through: threads, locks, instantiation, visibility, side effects that all relies on a central assumption: that you can own memory and direct it as you need with conditions. 

It functions on the basis that someone, somewhere, gets to build the thing and declare: this is it. And everyone else defers.

But Ethereum doesn’t do deferral. It does distrust.

There are no threads. No locks. No shared heap. No “global” anything. Every participant is in their own bunker, independently verifying what counts as real. You can’t enforce a singleton. You can’t even assume a shared timeline.

So your friend isn’t asking how to make a singleton. 

They’re asking how agreement emerges when every actor distrusts the others by default. How coherence appears when there’s no central authority to say, “this is the real one.”

And that’s what Ethereum is built to solve: not through protection, but through consensus. It doesn’t prevent duplication. It makes duplication irrelevant by structuring things so that everyone comes to the same result anyway, or gets left out of the chain.

Omg.

Is that what blockchain actually means? 

Blockchain doesn’t prevent lies. It makes lies pointless.

It’s not hype. It’s just a weird epistemological mechanism for incentivising convergence without control.

A protocol for agreement in the absence of trust.

A truth machine that doesn’t care who you are, only that you match.

Holy shit. How cool is reverse engineering? 

Back to the problem.

So, in Ethereum, singleton logic isn’t conditional so you can’t just VOLATILE and call it a day. Nor can you just eat the keys cause you’re a python, because that undoes the business model.

That’s where it becomes structural. The chain is the singleton not because it’s locked down by code, but because it’s handshook enough to lead to agreement.

That’s why the question isn’t really about Java or Python or C++. It’s about understanding what underwrites truth in a system that refuses to centralise. It’s about how we coordinate without consensus, and what is built to make consensus possible.

TLDR

"Explain how you would implement a thread-safe singleton pattern in a concurrent environment while avoiding the double-checked locking problem, and compare how this implementation would differ across Java, C++, and Python."

Answer:

In Java, the safest way now is the nested static class. It delays instantiation until the object is actually needed, and it leverages classloader guarantees to make that thread-safe without needing explicit locks. 

So instead of constantly checking if the instance exists (like in double-checked locking), it just builds it when the class gets used, structure enforces order.

No swings in face. Plenty of roof for all. 

In C++, from C++11 onwards, you’ve got std::call_once and std::once_flag, which let you initialise something once, across threads, without race conditions. It’s more manual than Java, but more transparent too. You get the sharp tools, you just have to use them right.

Do not use tomato knife for fish. (learnt that the hard way from an angry chef once) 

In Python, it’s weirdly “safe enough” most of the time thanks to the Global Interpreter Lock (GIL), which only allows one thread to execute bytecode at a time. But that breaks down in multi-process or async contexts, so if you actually want thread-safe singletons in those cases, you’re back to using locks, or reaching for external systems. 

Swings and knives everywhere, FIGURE IT OUT YOURSELF. 

BUT! Ethereum doesn’t work like any of these.

There are no threads. No shared memory. No global state. You can’t lock what doesn’t exist in one place.

So you don’t make a singleton the traditional way. You don’t guard the instance, you design the conditions so that everyone ends up agreeing on the same instance anyway.

Ethereum flips the singleton inside out.The chain is the singleton.

(Mean Girls comes to mind…THE LIMIT DOES NOT EXIST) 

Not because it’s protected, but because it’s ALLOWED by consensus.

And that, I think, is the real answer to your friend’s question. Not just how you do it in code, but what it means to do it at all, when trust is distributed and agreement is all you’ve got.

You gotta get those people speaking all different languages to make coffee, because if they don’t, someone’s gonna get grumpy. 

So Ethereum forces them to choose one, and pays them all for it. 

Hope that answers your question oh sphinx in the maze! 

More importantly, if any of this is correct Ethereum person: can I have a job? :) 

LAMBDA DISTRACTION SIDE QUEST

λ (lambda) is the 11th letter, but apparently the reason it ends up in programming has nothing to do with its placement — it's because of its role in lambda calculus, a formal system developed by Alonzo Church in the 1930s to express computation through functions.

This was one of the theoretical foundations of what it means for a function to exist and be computable. Like proto-math for software — before we had programming languages, we had lambda calculus.

Makes sense. Also why a lot of programmers (particularly the old guard) have a mathmatician style to gatekeeping and HATE vibe coders. 

I get it man, I feel the same about writing. 

In programming, a lambda function is an anonymous function — a function without a name — usually passed as a value to other functions. It lets you treat behaviour like data.

For example:

python

x = lambda a: a + 10

print(x(5))  # prints 15

Okay so what does this mean for concurrency?

In concurrent programming, you’re need to, or inevitably do frequently, pass little chunks of behaviour (functions) around:

• To be executed in separate threads

• As callbacks

• As tasks in queues

• As actions for futures, promises, etc.

And lambdas let you do this compactly, without writing full formal function blocks.

It’s lightweight, self-contained logic — perfect for handing off to another thread or process to run later. 

Makes shit faster to build, and less problems to run into, I get it. 

Also means that concurrency is coordination without centralisation. So, distributing tasks — possibly across threads, cores, or nodes — and you want:

• Small units of logic

• Clear boundaries

• Minimal side-effects

Lambdas are:

• Stateless or minimal-state

• Encapsulated (they know what they need)

• Composable (can be nested, chained, mapped, etc.)

So, concurrency adores them because they’re like smart courier pigeons: tiny, mission-focused, and don’t get in the way. 

Well, until they die on route. Lol.

Soooooooo: 

• Lambda in programming represents a compressed logic container.

• Concurrency is the practice of unfolding logic across multiple timelines.

So when you pass a lambda into a concurrent system, you’re doing something mythic:

You are dispersing meaning into fragments of time, asking them to cohere later.

It's symbolic delegation.

In distributed systems (like Ethereum), you can’t hold everything at once. You must pass functions of intent — little lambdas — and trust that coherence will reassemble on the other side.

So yes, lambda’s not just some quirky syntax. It’s a philosophical inheritance from the desire to compress intent, to express possibility, and to let execution happen elsewhere — in time, in space, in logic.

That’s kinda hot, glad I got distracted. 

Part 1. This is my Survival Document.

Otherwise, onwards and upwards.

The Myth of Intensity as Diagnostic

Much of what is currently named as “dissociation” is only treated as clinically relevant once it reaches a level of intensity that someone else can observe. But intensity is not a precondition of harm—nor is visibility a reliable indicator of significance.

This is one of the first conceptual cracks that emerges when clinical psychology tries to hold digital life to account: the assumption that what cannot be easily witnessed, diagnosed, or described in linear cause-and-effect terms must be either subclinical, metaphorical, or insufficiently severe.

But dissociation—like most survival-based cognition—is not defined by degree.

It’s defined by context, and by structural incompatibility. A person might “check out” during a conversation and be interpreted as bored or disengaged. Another might escalate into an argument, and be labelled defensive or volatile.

Both might, in clinical language, map to different trauma responses. But the difference isn’t in their intent. It’s in what the body determines is most likely to preserve the self within the given conditions.

The dissociative response doesn’t arrive with a moral or performative agenda; anyone who “performs” dissociation is clearly ill acquainted with how much energy it actually consumes. It is, in fact, the refusal of drama—an interruption of narrative in favour of pattern.

What happens is not the point. Why it happens, and how that “why” recurs across time, is where the point hides.

No one dissociates because they want to. Why would people describe the feeling as doomscrolling as “mindless”, if it were an active state?

We dissociate because the brain has been trained—by experience, by repetition, by the failure of logic to protect us—to prioritise coherence under pressure. When a current condition echoes a past condition that once required dissociation to survive, the response is automatic. The recurrence is the logic. Intensity is irrelevant.

This is why I find the emphasis on identifying and treating dissociation only at its most extreme—only when it impairs speech, causes full memory blanks, or results in prolonged detachment—deeply insufficient. The system that insists we must “feel disconnected enough” to warrant support is the same system that refuses to see the micro-adjustments we make just to remain legible in a space that has never accommodated our logic.

This is especially true in digital environments. When the body is not physically required, when input comes in fractured, overlapping layers, and when context is shaped by velocity rather than continuity, the dissociative response is not a crisis—it is how coherence is achieved.

This is what the clinical model misses: that dissociation is not always a detour from the self. Sometimes, it is the only viable route to staying with the self inside an environment that would otherwise flatten or erase it or render it at the mercy of whatever the algorithm decides regarding their existence.

And if the internet seems to provoke more of it, perhaps it’s not because digital space is damaging, but because it finally reflects the kinds of logic—recursive, indirect, structurally incompatible with real-time embodiment—that trauma-wired minds were already living inside.

We talk about trauma like it distorts or seeks to destroy a timeline. But it also builds its own. A timeline with new rules, new accelerations, and a new internal barometer for what pressure feels like. When that internal barometer recognises something as incompatible with continued coherence, it responds. Not with explanation. Not with moral clarity. Just with “this is what works, continue”.

So no—dissociation doesn’t have to “look like” anything.

It doesn’t need to interrupt the room. It doesn’t need to be validated by a peer, a partner, or a therapist. It is not an event. It is a logic. And until we accept that its appearance is entirely dependent on the environment in which it operates, we will continue to misread it as pathology rather than pattern.

Digital Survival Logic Is Not a Metaphor

When you stitch trauma-coded responses that have been generalised through arbitrary samples of “proof” across demographics, contexts, generations and into digital space, what you’re doing isn’t adaptation—it’s translation.

The core schema—fight, flight, freeze, fawn—doesn’t break. It drifts. It finds new coordinates in order to remain sensical to orientation. These aren’t approximations, they are reformulations shaped by an environment with entirely different demands.

“Freeze” becomes something like ghosting. Blocking. Vanishing. But not vanishing completely—just tactically. A soft erasure. Selective illegibility. It’s not hiding, it’s muting the surface without triggering a collapse of presence. You are still there, but only to yourself. That’s not avoidance. That’s pressure regulation.

“Fawn” stops being a performance for one person and becomes a pattern response to systems. You’re not appeasing someone—you’re appeasing a rhythm. You’re adjusting to algorithmic attention economies, to the tempo of feed refreshes, to the cadence of what gets surfaced and what gets skipped. The pleasing isn't personal. It's systemic. You’re not being submissive. You’re being legible. And for many of us, that is the only form of safety we’ve ever been afforded.

“Fight” loses directionality. There’s no target body to strike. So the strike becomes conceptual—trolling, moral warfare, subcultural in-fighting. The explosion needs shape, so it finds whatever friction can hold it. It’s not that people online are more aggressive. It’s that the fight response in an environment with no exit vector gets routed into recursion. That’s where you get obsession, spiral, shadowboxing. The target doesn’t matter. Only the momentum does.

And “flight”? It’s not about leaving danger. It’s about migrating to where danger hasn’t yet been mapped. The move from Twitter to Bluesky, Discord to Telegram, one groupchat to another—it’s not retreat. It’s architectural recalibration. The self remains, but the terrain gets swapped. You don’t disappear. This redirection is coded in platform logic: new username, new format, new self-state. No one asks why you left. That’s part of the design. That’s the freedom that makes re-entry possible.

The thing that breaks here isn’t the schema—it’s the insistence that it has to be somatic. It doesn’t. It can be procedural. Structural. Embedded in clicks, in silences, in how you navigate a comment section. A lot of trauma logic is pattern logic, and what is digital space if not a pattern engine? A space where logic lives not in the muscles but in the mouse movement, the scroll, the reply.

The clinical mind struggles with this because it wants to locate action in the body. But in digital space, action is distributed; scattered. You do not need to feel the survival response in the chest or the throat to be in it. You just need to be inside a structure that demands your body comply with its rhythm. And then you must observe the shape you’re forced to take.

If you grew up inside trauma, you already know how to respond to a system that’s always watching. That always asks you to perform the version of yourself that is least likely to be punished. Digital space didn’t cause the behaviour. It made it visible.

And it finally made it functional.

The Directional Self

There’s a recurring fear in psychological and philosophical takes on digital identity: that we are becoming thinner. That the self is being flattened, simplified, disembodied into two-dimensional avatars that resemble us only in outline.

I think it is not that we are dissolving or becoming less, we are acknowledging that human needs are not as simple as accumulating traits or habits or behaviours and routing them within ourselves, but it in movement between all states that can be possible.

So, it becomes directional.

It is not a loss of self, but a recalibration of selfhood to meet an environment in which continuity is not rewarded, but mobility is. And mobility—whether across tabs, platforms, voices, or relational contexts—is not dysfunction.

It’s fluency.

You are deemed inept if you cannot learn to use an email, not because you are stupid, but because it will encumber your personal experience with digital space, alongside everyone else’s. There’s no pressure to learn how to use email, but it’s understanding that if you cannot, it doesn’t mean that you need to prove that it exists in order to intentionally restrict you and others like you.

Where there are those who cannot untie themselves to the tether of who they believe themselves to be: Joe from Marketing, the person who has “been that way always”, themselves as an assemblage of chronological experiences and impressions upon them, to move like a cursor across pressure points, adapting its shape depending on where it is required to land.

It is not only understandable that something like email could become overwhelming, particularly with its frequency, its unpredictable context, the feeling that comes with it that it is a letter sitting in a mailbox, and that the longer it sits, the more it deteriorates and you will be responsible for it. That the communication feels enough like a conversation that it follows the same rules, that they can repeat, remind, you can be reprimanded for missing or misinterpreting.

There is a physiological response to being denied you exist when you have been taught to sit still then asked after 50 years to get up and dance.

People confuse this movement with superficiality. They think a directional self lacks depth because it doesn’t accumulate in a single, narratively consistent identity. But that’s a nostalgic reading.

The self online isn’t reduced—it’s accelerated. Routed through expectation, context, platform constraint. And those pressures require a self that can change direction midstream without collapsing.

If anything, the directional self is under more pressure than the fixed one ever was. It has to signal intent without context. It must be simultaneously responsive and anticipatory.

Not because that’s what the person wants to be, but because it’s what the environment demands.

You either learn how to shift or you get misread.

Depersonalisation Is Not What You Think It Is Here

Depersonalisation isn’t the failure of identity. It’s the cost of having to manage its expression across incompatible terrains.

If you do not know where you are supposed to be on the internet, and do not have the tools to understand how to move within it, of course you get stuck in spaces that make you feel unwell.

This is the scourge of corporate spaces on the internet, in particular social media. Entrapment within spaces, providing the simulation of movement, but really embed the belief that you need to be there, specifically there, to exist, to be relevant? That’s the root of the illness, a demand that

And this is why I resist the language of “flattening.” What looks like a reduction from the outside is often a form of compression. A strategy. A way to transmit the maximal signal through a limited bandwidth. It’s not a lesser self—it’s a self under different structural terms.

We don’t describe actors as fragmented for being versatile across roles; the opposite is in fact true. We don’t pathologise bilingual speakers for switching codes depending on audience. But we still treat digital modulation as suspicious—as though the only valid identity is one that stays still, says the same thing, and does not alter shape.

But for those of us raised inside trauma-patterned logic, altering shape is how we survived.

The digital didn’t invent that logic. It holds in a space where the pressure to remain continuous is finally surpassed by the need to remain legible.

If the physical world demands that we “be ourselves,” but offers punishment when that self fails to meet norms, then the digital world offers a different proposal: be a self that moves. Be a self that learns. Be a self that understands that direction is not deviation—it’s design.

And for some of us, that’s not performance. That’s relief. The rest, well, I believe that is the physical worlds expectations that are required to answer for the illness it has caused as result.

Platform Logic and the Illusion of Autonomy

It is a kind cruelty—though cruelty nonetheless—that platforms are built to suggest freedom while demanding compliance. You scroll because you want to. You post because you chose to. You perform because you’re being authentic.

Except you’re not. You’re obeying.

And that obedience doesn’t look like constraint—it looks like coherence.

You are free to express, but only within forms that map easily to metrics. You are free to change, but only if the change can be read as evolution. You are free to disengage, but only if your absence can be understood as a pause, not a failure.

Platform logic doesn’t ask you to be yourself. It asks you to render yourself—smoothly, legibly, and in terms it can process.

This is what I mean when I say that the anxiety people feel in digital space is not about “losing their sense of self.” It’s about the cognitive dissonance of navigating a structure that demands fluidity while punishing inconsistency. You don’t get to choose what parts of you are rewarded.

You just learn what gets seen, and that is what you become moment to moment, or, over and over.

We mistake this dynamic for self-commodification, as though the individual is uniquely at fault for treating themselves like a product. But the conditions predate the choice.

The user is not commodifying the self—the platform is. People learn “valuable” skills in order to excel at a job, what is the difference in an environment where visibility is the goal, and the valuable skills whatever leads to it?

No one is “becoming”. It’s the same sort of thing as “becoming” your job. I think of how when people introduce themselves and the first question to follow “how are you?” or their names, is “what do you do for work?”

You are described by the way you are useful through the lens of economic viability. You're being assessed through the vignette of how you do, or do not make money. And how much? Why? Etc.

The example made in the article is dating apps, but this is where interface design explicitly reduces identity to components: age, face, tagline, location. You’re slotted into fields and then blamed for becoming field-shaped.

If you pay, you get more options.

If you don’t, you’re filtered into a lower strata of desirability. This isn’t social failure. It’s algorithmic triage.

Instagram. LinkedIn. Substack. Patreon. TikTok.

Every platform functions on the same logic: render the most legible, desirable, and interaction-producing version of yourself—and you will be rewarded. But only while that version remains current.

Being fired because of “who you are” on Instagram, is like being fired for being a character in a play, compared to the one that goes home, showers and unwinds before bed.

This is the illusion of autonomy that underwrites the entire digital experience: the idea that your ability to shape your selfhood is personal, moral, or expressive, rather than infrastructural.

You’re not making a brand. You’re following instructions—some of which are visible, and some of which are felt only in the nervous system, when posts don’t land, when a feed grows quiet, when a shadowban is suspected but can’t be confirmed.

Seen = safe. Seen = valued. Seen = real.

And the cost of that visibility isn’t just emotional. Because the more time you spend translating yourself into terms that a platform will understand, the more estranged you become from the version of yourself that existed before the translation.

What if the platform isn’t distorting anything, but simply revealing the modularity we’ve always relied on, now rendered as interface?

What if this sense of dissonance is not the result of digital distortion—but of finally seeing, unshielded, how contingent identity has always been?

This is the crack the article never quite touches. It is avoided often, because to admit so, particularly where the stakes are looping your professional persona in with the one you’re occupying online, are fraught.

In psychology, they always get close—acknowledging dissociation, disembodiment, digital ego states—but ultimately retreats to a therapeutic model that frames the loss of self as a problem to be fixed. It presumes that autonomy exists in some natural, pre-platform state. That fragmentation is a deviation. That the task is to return.

But what if there is no return?

What if the integration they long for was never a real starting point, but just a story that worked for a while—until it was outpaced by scale and then, by showing there was not one shape to hold structure?

Algorithmic Pressure as Directional Force

What is often mistaken for anxiety in digital space is not the existential loss of a stable self, but the continuous recalibration of the self within structure.

Algorithmic pressure and its effects are specific to the intrusion of standards that existed before digital spaces were created, and in the early days of digital spaces when it was still being “felt” out.

When people say they feel “lost” or “tired” online, what they are describing is not disconnection—it’s the strain of maintaining coherence under conditions that demand constant adjustment. This is what algorithmic pressure causes, through its artificiality in an environment where nothing is artificial - everything exists.

This isn’t about knowing who you are and forgetting because of over-exposure and adaptation, or even a flattening sense of self. It’s about being asked to become, again and again, under architectures that do not allow for embodied rest.

That ask you not simply to exist, but to format yourself in a way that can be measured, traced, and rendered interpretable to others—and to the machine.

That is not flattening, dissolution or even disorientation, that is contortion.

The algorithm is not a force that notices you and rewards you. It is a force that notices the shape of coherence as has been programmed into it. Once it is identified, it accelerates it. The closer you are to a recognisable rhythm that the algorithm is designed to capture, the more visible you become. But visibility is not attention. It is demand.

Every time you engage, you feed it. You confirm its predictive accuracy. You refine its model. And that model becomes the terrain you must now navigate—until your past actions become the only route forward. It is not memory; it is momentum that only applied to shapes that fit the slipstream.

And the longer you’re there, the harder it is to remember where you began.

This is not dystopia, a barren terrain to navigate or one that is a living trip wire.

This is recursion.

What makes trauma-wired or digitally native minds particularly susceptible to algorithmic influence is not their fragility—it’s their fluency. These are people whose cognition is already recursive, already patterned, already oriented around velocity as a form of safety. The platform recognises that movement. But it doesn’t follow it. It understands it as something that if rewarded, it is in a better position to manipulate.

Recursion, when enacted by the trauma-wired mind, is a method of mapping. It’s self-directed, even when disoriented. It spirals not to collapse, but to recalibrate. It repeats not to regress, but to locate.

But algorithmic recursion is not spiral; it’s compression. It performs pattern-seeking not as meaning-making, but as exploitation. Not “you are here again, what do you see differently?” but “you were here before, and we will bring you here again until you perform what we consider legible.”

This simulated motion—this loop that appears like a spiral—is where the danger lies. Because to the trauma-shaped mind, pattern equals survival. But when the pattern is governed externally—when it is enforced, predicted, and refined by an architecture that uses your pattern logic against you—it stops being survival. It becomes erosion.

This is why the algorithm can feel uncanny, even to those fluent in its structures. You recognise its movement, but not its intent. You see your own logic mirrored—but with the agency drained out. You sense the velocity, but not the choice.

For someone whose sense of self is already distributed across survival states, the pressure to remain visible within the loop—to simulate your own motion just enough to stay legible—can feel like dissociation, not from the self, but from motion that no longer belongs to you.

The harm isn’t that digital space moves. The harm is when your native movement is co-opted by a system that pretends to move like you, but doesn’t allow you to change direction.

This is Coercion, Not Freedom.

This is why consent becomes so distorted in these spaces, and why attempts to honour traditional standards of consent, do not translate.

E.g. harm minimisation by punishing hate speech, by banning teenagers from social media to control its effects on the psyche or by putting a paywall or an age lock on a website to acknowledge “only this person is allowed”.

The platform doesn’t spiral with you. It loops. It narrows. It optimises. It says: if you reacted once, you will react again. If you paused here, you will pause again. If you opened this, you will want more like it. If you hesitated, we will replace your hesitancy with speed.

This creates a situation where survival becomes a matter of prediction management. Not resisting the algorithm, but appearing predictably unpredictable—enough to be interesting, not so much that you vanish. Enough to disrupt, not so much that you collapse.

That pressure is not emotional. It’s structural. And so is the cost.

Because the self you upload is a self formatted for legibility, not complexity. A self shorn of the unfinished edges. A self that responds cleanly to prompt, that doesn’t glitch, that doesn’t ask for a moment of incoherence to regroup. And the longer you spend in that state, the less accessible the unrendered parts of you become.

And yet those parts still act. Still think. Still twitch in the body. They just no longer show up in the feed. This is where the gap widens between what is seen and what is lived. Between what is presented and what is endured.

The algorithm doesn’t kill the self or split it. It trains it. And like all training, it does so through repetition and reward.

So, when the article frames this as a “trance state,” or suggests that dissociation online leads to flattening, the assumption is that disconnection is the problem. But for many of us, disconnection is the skill. It is what kept us alive in the environments that asked us to be too many things at once. The internet did not invent that. It just gave us a place to do it without immediate bodily risk.

This is the difference between reading algorithmic pressure as oppression versus reading it as gravity. Gravity is not bad. It’s just a force. But when you’re only taught to walk in one direction, any other movement feels like failure.

That’s not the internet’s fault. That’s the lie of a society that only recognised a narrow range of stability as legitimate.

And here’s the paradox: the same structure that constrains us is the one that allows for recognisability. The algorithm mirrors trauma not because it harms us in the same way, but because it structures interaction according to survival logic—repeat what works, abandon what doesn’t, follow the pattern until it breaks.

To ask people to ignore that logic in favour of integration is to ask them to discard the one form of intelligence that keeps them alive in an environment they are expected to understand as it inhabits every space we occupy, structure irrelevant.

So we don’t. We comply, recursively. Not because we are weak or addicted, but because compliance is not moral. It is directional. We do what moves. We stay where the pattern holds. We go quiet when there’s no thread to follow.

And if you want to know why some people don’t “log off,” it’s not because they can’t. It’s because this is the only terrain where their movements have ever made sense, or they have trained that this movement, will allow them to maintain the concept of sense itself.

Integration Is Not the Goal

There is a growing split—not between those who are online and those who are not—but between those whose cognition aligns with the structure of digital space, and those whose presence online is forced through the framework of physical-world expectations.

The first group is not necessarily younger, or more tech-savvy, or “chronically online.” They are those for whom digital space functions not as an escape, but as a compatibility layer.

Their mind naturally, or can move, like a meme: recursive, relational, non-linear, non-coherent by design. Their digital presence is not fractured because they are lost. It is dispersed because that’s where fluency lives.

These are the users who intentionally become incoherent—not as rebellion, but as rhythmic refusal. They drift. They distort. They operate multiple selves across a single platform, not to mask instability, but to show that motion is coherence. The same way a meme changes its shape but not its function. The shape-shift isn’t deception.

A meme shows you exactly how the motion of the internet “works”. The ones that move and the ones that don’t, how they resurface, how they distort - all evidence of motion that is natural to the internet.

Contrast this with the second group: those who are in pressure cookers of physical, institutional, or familial coherence, and who are forced to carry their “whole self” into digital terrain.

They are told to be visible, but not too much. Consistent, but not robotic. Authentic, but not weird.

These are the ones most at risk of collapse—not because they are weaker, but because they are asked to perform legibility across two incompatible value systems.

This is the condition that produces what we now call "influencing." Not as aspiration, but as a compliance schema. You become a brand not because you want to sell something, but because the only viable identity left is one that performs the illusion of static coherence across time and space.

Notice how lip filler disappeared with the Kardashians? Interesting hey.

Notice how the Kardashians disappeared full stop? Not their intention, but the movement of the internet decided for them. They will call it intentional, or fatigue, it’s not. It’s the registering that you have worn out, and your brand hinges on the lack of contortion you go under and the narrative strength of your being. These things are increasingly unimportant in the saturated feed.

You optimise not to grow—but to survive in both terrains. And when the logic of each space contradicts the other, what do you do? You become a symbol. Not a self. A product legible to the algorithm and reassuring to the family, the peers, the school, the feed.

But here’s the lie at the heart of that model: that the self should be integrated. That it once was whole, and digital space has split it. That fragmentation is the wound, and integration is the healing.

But what if the integrated self was always an aesthetic, imposed by people whose conditions never required splitting in the first place?

To survive trauma, we split. To survive institutions, we perform. To survive online, we shift. Which of these is not integration?

Or more precisely—which of these integrations are counted?

“Integration” in the clinical sense assumes that stability is found by returning to a prior self, or becoming one that is coherent, steady or constant. But for those of us wired by structural rupture, the “before” self was never fixed.

It was a survival artifact.

The digital, at its best, didn’t destroy that—it exposed it. It showed us that identity isn’t a container. It’s a rhythm.

It’s why there are people online who are culturally “significant'“ to the digital age but feel almost uncanny, and those who are who are significant who feel shapeless and almost un-human.

Digital Native Influencers Versus The World

A good example to work through that previous section, is made through influencers, one in particular comes to mind.

Abbie Chatfield represents the performance of adaptiveness, not actual digital fluency. Her trajectory mirrors the application of a legacy media model onto digital space: linear career progression, media training, brand partnerships, controversy management.

But all of that is built atop the assumption that visibility and authenticity are compatible.

They are not. Not anymore.

What’s clunky isn’t the content—it’s the friction between what digital space rewards (velocity, recursion, illegibility-as-relatability) and what traditional public figures were trained to manage (consistency, clarity, brand coherence).

So what you end up with is a simulated messiness, packaged as “real,” but following a script that assumes moral resolution is not only possible but profitable. The digital sees through that instantly.

Whereas the truly fluent ones—especially on TikTok—don’t simulate chaos, they embody the logic of drift. They don’t build personas to withstand scrutiny; they build formats that can mutate. They are not “relatable” because they show us their lives.

They’re relatable because they replicate how it feels to be online.

That surreal, rhythmic, pressure-cooked drift where the joke is a coping mechanism and the dance is a refusal and the caption is a second-order recursion of the thing the body cannot say.

Abbie Chatfield is the image of adaptability, stilled into branded coherence. TikTok trend-creators are adaptiveness itself, pulsing and distorting with the pressure of their moment. They’re not building selfhood to survive—they’re shaping the structural terrain that survival now takes place in.

When people like Abbie Chatfield appear clunky online, it’s not because they are bad at the internet. It’s because they are performing the wrong logic. They are applying linear rules to an ecosystem that runs on acceleration, recursion, and the distortion of the self as form, not character. Their adaptability isn’t movement—it’s pre-approved shape change.

TikTok trend-makers, meme artists, shitposters who loop through irony, glitch, hypertext, absurdism, refusal, and multiplicity—not because they want to obscure who they are, but because their presence is structurally relational. Their identity isn’t in what they say. It’s in how they move.

Digitally fluent people aren’t trying to look “whole”—they’re trying to be readable at speed. You’ll see accounts with 5 usernames, cryptic bios, layered irony. They're resisting capture while still remaining visible.

That’s not disorganisation. That’s defense architecture. It’s also the most human thing a person can do when under pressure: modulate presence, control legibility, reduce the surface area of vulnerability. The irony is a smokescreen; the glitch is camouflage.

Compare this with Abbie Chatfield-style branding, where “vulnerability” is confessional and flattened into digestible, emotional coherence: I was like this, but now I’m this, and here’s how I got here. It has a beginning, middle, end—and a punchline. Digital-native incoherence has no arc. It loops. It pivots mid-sentence. It vanishes. It’s not hiding—it's digesting.

TikTok creators don’t think of themselves as people, it is a program you run when you need to. A dance isn’t a dance—it’s a file you run through your own body. A skit isn’t a joke—it’s a reference engine that gains meaning only in collective rhythm. The form moves through people. That’s fluency.

It’s also the opposite of the legacy influencer model, where the person is the product and the format builds around them.

Influencers with real-world logic apply the formula backwards:

I exist, so the platform must accommodate me.

But digital fluency says: The platform exists, so I will learn how to move inside it without fracturing.

It’s about survival within logic, not despite it.

Fluent creators aren’t apolitical for this reason—they’re structurally evasive. They move through and alongside contradictions, not to resolve them, but to avoid being trapped by them. They use queerness, irony, absurdity, refusal, hyperbole, stimming, repetition, glitch—all to indicate that they’re inhabiting a space that can’t be made moral through narrative closure.

This is why so many "media trained" influencers are perceived as disingenuous when they make “statements”—because they're applying narrative closure where people are already fluent in recursion. It's not that users want no accountability.

It’s that they want structural coherence before they want moral performance.

Watch how people move on TikTok versus Instagram versus Twitter (X) versus Discord. Digitally native people don’t “post.” They ride the logic of the tool. That’s why TikTok creators feel alien when they jump to Instagram—because IG rewards static aesthetic cohesion, not adaptive motion.

Abbie’s clunkiness, or any influencer like her, are not a personal failures. They’re the artefact of misapplied logic: moving linearly in a space that no longer believes time is real. Her apology videos, her “hot takes,” her “raw” talks—these are all attempts at coherence in a structure that rewards fragmentation performed fluently.

And this is where value becomes apparent, and the means.

Some monetise because they must, others because they can’t not. The difference?

• The first group commodifies presence and coherence.

• The second group commodifies velocity and pattern.

Fluency is not about being anything. It’s about surviving the structural conditions of a space that was never designed to reflect back a whole self.

The fluent ones know this, which is why they keep changing shape. The others try to restore shape, and that’s why they clunk.

Digital space didn’t break the self. It revealed that the self was already held together by performative rituals: dress codes, tone of voice, office etiquette, school uniforms, facial expressions that count as “respect.”

When those rituals fall away, you don’t lose the self—you see it. It was always recursive. Always in revision. Always under construction. You were just too busy holding it together to notice.

Integration is not a goal. It’s a constraint. And when you build a theory of the mind around integration, what you’re really doing is demanding stasis. You’re saying: hold still so I can understand you. Hold still so I can name you. Hold still so I can return you to the system that needs you to stay in one place.

But for those of us who only survive through movement, that’s not healing. That’s erasure.

To demand integration is to ask a recursive system to halt. But systems like us don’t halt. We cycle. We shift. We change shape mid-sentence. And the digital finally gave us a terrain where that wasn’t pathology.

It was the interface.

It is not flattening—it is the recognition that the self we imagined was only ever held together by demand. That we are seeing that the self is not what we expected, and that is an ego death, not a state. One that is required and doesn’t require acknowledgement or dissociation to comprehend who one must be, only due to demands of an artificial structure.

It’s not pathology. It’s structural readjustment.

The System Is the Disruption

The disruption is not digital space. The disruption is the expectation that there was ever not disruption.

Clinical language, institutional logic, and even a large portion of media critique still carry a baseline assumption: that once, people were integrated. That there was a time of selfhood, of linear cognition, of moral coherence, and that the digital has disturbed it.

But for those of us wired through rupture—neurologically, structurally, socially—there is no “before.” There is only structure.

The idea that digital space introduces fragmentation is a projection from those whose coherence was never tested until now. Those who were made legible by the world as it was, and are now demanding the internet re-perform that recognition. But for the rest of us, digital space isn’t a disruption. It’s a system that finally reveals the others were always fragmented too—just better camouflaged.

The internet did not create trauma logic. It didn’t force multiplicity. It didn’t fracture the self. It stripped away the scaffolding that made those things invisible. It revealed the recursive, mimetic, dissonant, nonlinear architectures that most people had spent their entire lives suppressing through performance.

This is why people feel “crazy” online. Not because the space itself is harmful—but because it holds up a mirror and shows you the incoherence that was always there. And when there is no familiar script to resolve that, when your identity is no longer protected by the rituals of physical space—etiquette, fashion, tone, hierarchy—you’re not disoriented.

You’re exposed.

This is not a problem to be solved. It is a structure to be understood.

This is where even the best-intentioned writing about trauma and the internet falters. Because it keeps trying to fix something that is only broken if you assume your static standards are the goal. But for people whose lives have never allowed for that kind of stability, the presence of a system that functions despite incoherence isn’t disruptive.

It’s affirming.

Visibility At Scale.

The clinical mind wants to identify dysfunction so it can restore function. But what if “function” has always been an aesthetic? What if the real harm isn’t in multiplicity, or fragmentation, or recursion—but in trying to trap a moving cognition in a static container?

What if the system was the pathology all along?

What you name “symptoms” are often structural adaptations. What you call “maladaptive” is often rhythm tuned to a different drum. What feels like suffering is often the dissonance between how you move and what the system demands you become in order to be witnessed.

And here’s the hardest truth: the digital didn’t invent that system. It just removed the padding. It removed the costumes and scripts and physical buffers that allowed people to pretend their legibility was natural, not enforced. Now, without the buffer of physical performance, the incoherence becomes visible.

Not created. Visible.

And this visibility? It scares people. Not because it’s new. But because it proves that the “normal” self, the “coherent” identity, the “integrated” psyche were always performative conditions. And that means everyone is closer to collapse than they thought. Or more precisely—they were never not collapsing. They were just performing balance on a loop.

That’s the disruption: not trauma, not speed, not overstimulation.

The sudden, recursive rendering of how a self is made and sustained, the reveal that “being okay” was never a state—this is just a mask worn convincingly, and the mask doesn't work in code.

In this space, the demand to integrate, perform, or stabilise becomes not only incoherent, but cruel. Because it denies people the right to exist inside the motion they were built to survive.

If you feel exhausted by the internet, ask yourself: are you tired of being online—or are you tired of performing recognisability in a space that keeps asking you to prove you exist? Because the internet doesn’t ask that. The system does. The system always did.

Digital space just makes that demand seen.

Deep Breath Now.

Digital life is framed as an interruption, a distortion, a less-than-real proxy for the embodied. As if embodiment itself wasn’t also a performance under previous conditions. As if the body has ever been a fixed guarantee of presence.

In a world that constantly argues the conditions that make you matter? What makes you think that the internet would behave any differently with those standards applied?

When we speak about the self in crisis, we rarely ask: whose crisis? According to whose standards?

The internet didn’t make people incoherent. It gave incoherence a form. A terrain. A rhythm. A logic. And for those of us shaped by trauma, by dispossession, by cognition that defies the linear—this wasn’t a disruption. It was recognition.

We are not trying to integrate. We are not trying to stabilise. We are trying to continue. And sometimes, that means changing shape so completely that nothing familiar remains except the movement itself.

Through this, we see digital space as a fault line that ached and groaned until the erosion said, you know what, it’s time.

It is collapse caused by the system. A structural rendering of all the conditions that made performance mandatory and made coherence a virtue. And in that hole it’s leaving, we see not dysfunction—but design.

A design that no longer fits.

This is why the language of pathology fails here. Not because people aren’t suffering. But because the framework is built on a definition of “self” that assumes integration is both desirable and possible. For many of us, it’s neither. We’re not trying to return to wholeness. We’re trying to survive structure.

And when survival requires drift, recursion, multiplicity—then those aren’t symptoms. They’re tools. Not signs of illness, but maps. O

To pathologise this is to demand that people amputate the very strategies that keep them alive.

So no—this isn’t about flattening, or detachment, or disconnection. This is about re-tuning what it means to think. To appear. To mean something in a space that recognises your presence only when you fit—and punishes you the moment you fall out of sync.

But if you’ve made it this far, you already know: the sync was never real.

The sync was a performance. The sync was imposed. The artificial sync that you were asked to find is was what got you here. And now, you’ve realised you followed a voice not in your own head, but all around.

We move because we need to, and the system asked us to be still.

We just got our circulation back.

I really didn’t intend it to be this long - oops. But don’t be cry, it is okay.

I separated it into pieces so that the scope is understood, that the length is not for the sake of it, but rather to demonstrate the recursion, the mapping and the cognitive cartography of which I now know is innate to my thinking, and thus, my writing.

Here are the contents so you can CTRL-F if you are interested in certain parts.

Piece One (In this post)

• First, Context.

• Why I Am Similar to the Internet (and maybe you are too)

• How I Experienced Being "Clinically Unwell”

  • Social Difficulties

  • Changes in Speech

  • Difficulty Learning

  • Difficulty Concentrating

  • Dissociative States

• My Drugs, And My Therapies

  • EMDR, DBT, ADHD, Lamotrogine.

• Being Allowed To Think.

Piece Two (link here, or at the bottom of the post)

• The Myth of Intensity as Diagnostic

• Digital Survival Logic Is Not a Metaphor

• The Directional Self

• Depersonalisation Is Not What You Think It Is Here

• Platform Logic and the Illusion of Autonomy

• Algorithmic Pressure as Directional Force

• This is Coercion, Not Freedom.

• Integration Is Not the Goal

• Digital Native Influencers Versus The World

• The System Is the Disruption

• Visibility At Scale.

• The End.

This is the piece I am writing from, here:

Next Level Psychology

The Age of the Divided Self

How the Self Shatters…

Read more

2 years ago · 21 likes · 6 comments · Jeremy

I want to make clear that I do not take issue with the author’s framing.

In fact, I think it’s a generous and grounded attempt to navigate territory that’s increasingly fraught. Its therapeutic, and offers something valuable: a language through which people might come to recognise the altered states, recursive selves, or dissociative tendencies that arise under pressure, especially in environments that seem to fray coherence at the edges.

But I am not a clinician. Nor am I interested in therapeutic restoration.

I have no stake in helping people return to an integrated self—because, for me, the integrated self was never an available baseline. Instead, my mind is one shaped structurally by complex trauma, and made legible to me only through digital space, not clinical psychology.

First, Context.

If I can become what is considered clinically “well enough” to keep myself alive in the physical environment, embodied as the standard expects, I want to express that this is not because I have conformed to static expectations, values, behaviours or a moral code that allows that to be so.

Through digital spaces I have been able to recognise and map the structure of my thinking, not changed, but learnt how to survive it.

I do not view digital environments as abstract disruptions of the “real,” but as the first architectures that rendered my way of thinking not only functionally possible, but intelligible to anyone or anything, other than myself.

As I improve, it shows that one of the reasons I continue to, is the pattern seeking that my brain requires to soothe, is satisfied with the amount of information it receives. With this information, I know when to evacuate me or allow me to remain. Not perfectly, but more so.

But now, I’m seeing an inversion of what I grew up with trying to understand.

I am encountering more and more people who are experiencing the implications of digital space movement. It is a systems error, one between a space that requires you to think as a trauma-wired person would and the structure experienced by all pre-digitally. This is fracturing harder, faster, as the pre-digital world forcefully presents itself through every avenue it decides to access.

This is not saying that people are being traumatised by these implications, as is so often described. I think that people identifying it as that are often incorrect, and this labelling is harmful to those who live within trauma logic as result from specific events and the resulting forced compliance, rather than digital exposure.

This isn’t a failing of individuals, who are seeking to name what is unsustainable for a psyche or a self.

It’s a failing of the values of what we associate as well and unwell, and that generalising cognitive responses like anxiety or dissociation, and linking them with mental illness, has only proven the limits of clinical psychology as they stand.

Again, not an individual failure of clinicians (power to you guys, it’s hell out there), but that they have been confronted with a scale of distortion they cannot hold as they are currently structured.

Why I Am Similar to the Internet (and maybe you are too)

Note: I was not born in 2000. I was born in 1992. A digitally insane millennial. Believe it.

One of the things that people do to relieve themselves or force dissociation when they are under extraordinary pressure is engage in drinking or drug taking. Like me in the before times. (421 days sober baby!)

Relapse is so looming, always, because when encouraged to reform, you often return to conditions that are not only environmentally precarious, but you are also mentally and physically unprepared to work within in order to be considered “well” or useful to society.

Most of these people are traumatised as children, not in adult life, so the soothing tactics of alcohol that cause effects after the fact, become the focus or the “locus” of dysfunction.

The fact that say, you have difficulty with speech due to the neurological impairments alcohol has caused, and this makes you unable to explain yourself when struggling with a task, becomes statistically a more realistic outcome that you are fired or held back, is illogical. It is apparent when a person who could, or had a speech difficulty through not coping mechanism, but as an acceptable, societal reason, would have different stats to you. So, less like to have suffered the same consequence.

This is where trauma comes into the conversation, and its use across those who are experiencing disharmony with digital space and physical space clash.

People are showing prolifically and ubiquitously what are considered "unwell” behaviours from a clinical psychology perspective. Social difficulty, changes in speech, difficulty learning, difficulty concentrating, or what’s described as prolonged dissociative states.

What is being described as trauma is because our current cultural language lacks the nuance to differentiate between structural overwhelm, sustained environmental pressure, and the aftermath of rupture that rewires the nervous system permanently.

You can remove the alcohol, but you have not removed what caused you to drink in the first place.

This miscategorisation is not benign. There is a reason we are all fighting over categories assigned to everything we see, in order to sort and keep “selves” legible amidst this distortion.

When every fluctuation of stress, derealisation, or dissonance is framed as trauma, it becomes harder to see the structural patterns that trauma actually creates—and even harder to see when a person’s cognition has been shaped by those structures, not temporarily disrupted by them.

At the same time, this surge in self-diagnosed and ambiently diagnosed distress has exposed that the systems built to contain and interpret individual pain are now being asked to process collective collapse. It is not working.

The language, the metaphors, the metrics of wellness, all begin to falter under the weight of so many people expressing what used to be pathologised as an exception, and a thing that must be fixed or the “self” will be lost.

How I Experienced Being "Clinically Unwell”

Now, let’s provide some context as to how I experienced “unwell” behaviours as a teenager and through my twenties, so that there is enough to flesh what I will address afterwards.

I want to express extremely clearly, that this is not making light of my mental health journey. I am expressing this informatively for context of the piece, this is not a memoir - because I do not need you to understand or feel any of the pain I went through because it is not the pain that is the issue.

It is the structure that enabled me to feel it specifically and so intensely in these ways.

Social Difficulty

After a traumatic event, I found myself deeply avoidant and hyper-critical of women, and drawn to men not out of respect, but fear of being abused again. For me, proximity was a reliable way to acquire knowledge to understand how I could keep safe.

The logic:

This is what men do when X, so I know that the response I have to that is Y, I will continue to do Y to make sure that X happens.

Women on the other hand I deemed as weak, volatile and susceptible to damage; understanding I was a woman then and that this happened to me, because I am one. I’m sure the fact I know I am non-binary now is also part of this, gender dysphoria very clearly something lassoed by ideology but missing the point that gender and sex are constructs, only function is to categorise and sort people by function.

That sidebar could lead other places, but let’s remain here. I also didn’t understand the lack of care from female figures I’d come to associate with providers. That is not a condemnation of them, that is an interpretation of my own personal stress of incompatible cognition with the world around me.

Instead, I received judgement for my symptoms of traumatisation - those inclusive but not limited to disordered eating, inability to conform to social expectations and a deep misunderstanding of what it meant to alter my image with makeup, hairstyles or clothing.

That last one because sensorial inputs like necklaces felt like choking, and changing my face or hair resulted in me not recognising myself, and when I saw it in others, not recognising them either.

Changes in Speech

I do this impulsively. I will mirror the tone of whatever the person who is talking to me, often accent as well. This is more than just chameleonic, or a people pleasing reflex, it is a “sorting” of self into one that a person will interpret as familiar or safe. You mirror first, because it is the easiest route, and often with the least resistance.

It’s often why people who get to know me notice I have completely different voices or “characters” (by voice, not personality) depending on the context I’m in, what I am talking about, and who I am talking to.

I find in moments of high distress or confusion, I will immediately change my tone to something monotonous and almost unmoving. This is an attempt to “refuse” communication, signal - I am not ready to interpret anything else other than the distress I am under.

I very, very rarely, if ever, would be entirely silent. To me, that is overwhelmingly proven by pattern that silence, will be unsafe for me. I will not be able to measure reaction. I will not be able to assert my existence. I will not be able to understand spatially where I am.

It’s not that it is unsafe, it’s that my brain registers it as true and unmovable unless proven by pattern otherwise.

Difficulty Learning

I have been described as a savant, but only when I choose to be. Because my brain works relationally, I do not retrieve memory in a linear fashion. Yesterday is today, tomorrow is four weeks ago, I was born but I cannot understand how or when. This is normal in CPTSD because your re-wiring includes a cool functionality for memory in that it rejects the calendar and the Western concept of time.

I will learn voraciously (just look at the subject matter across my Substack for a microcosmic example LOL), but only when I am free to do so within the requirements of my thought process. E.g. on Substack.

So, when I attempted university, I ended up dropping out three times. At university, I once wrote an essay that had a question prompt, and because I finished it close to the due date, I looked at the question and then wrote, not considering I would need the whole essay to stay on the rails of what the question contained, and instead wrote something that sprawled across ten different subjects, and finished continents away from what was asked of me. I could not understand what was wrong with it, I suppose I do now in the sense, I didn’t do what I was asked.

But it seemed the only logical way to complete it at the time.

Difficulty Concentrating

See above.

Also, I will find it impossible to track a conversation on the phone, because there is no context. I usually won’t remember anything that’s said, or that we spoke at all. Written communications, I will remember for the rest of my life - particular threads and group chats online that jump and skip like thought does.

I will often listen and answer, but sometimes I am answering a question I think could have been asked before or afterwards, disorienting whoever I’m talking to.

Sometimes I will interrupt (well, all the time) not to insert my opinion, but to either repeat what the person has said in order to confirm they have said it, or because I do not understand what they have said, so I have offered it again so that they can confirm that’s what they have said to me. Sometimes I repeat to show that I am no harm to the person who is speaking, that I acknowledge what they have said.

If it’s instructions, God help you trying to relay that to me - it will take longer than anything, just write it down please.

Just don’t expect me to complete them on time. Unless that too, is written on there.

:)

Dissociative States

This will be when I can no longer place people, my memory tends to dissolve, and I tend to get in patterns where I binge watch TV shows, play games that are cosy for large blocks of time, or find it impossible to “begin” a task not because I am lost or frustrated, but because there is no logic to follow.

No logic, no pathway, no movement.

I used to compulsively lie when I was a teen. Not because I was vindictive or manipulative, but because:

a) I couldn’t remember anything that happened so I would say things as I needed them to be true in order to make the conversation coherent or move forward in a way that relieved me from danger.

b) I didn’t hold attachment to any information I received or transmitted, everything said to me was a lie or a truth, so my output simply matched it.

I would stand in front of a mirror and push my face together and stretch it, in order to see the parts of my face that I would “recognise” in photos. I would imagine this with others I had in my life, so I could do the same to “recognise” them.

In terms of more classic dissociation, I’ve never had a sense of self I can locate, it was always dispersed across whatever context I required to be in. Psychologists have often fired me, because they think I am playing a game and do not want to be well.

It’s not exactly that, it’s just that, I simply do not care to locate it, because structurally, a static and contained self, is what harmed me in the first place.

Also because there is not enough in the liminal space of a therapists office to calculate, so I have to do it through conversation. This, leads to interesting conversations…

Again, this is not because I was manipulative, but because I was shown through pattern that in order to survive certain contexts, I needed to be a certain way.

Hospitality as my career for the past fifteen years (thankfully, now over) put this into hyper-speed, and made it loaded with financial stakes alongside my own personal ones otherwise. Become who you need to table-side, otherwise the goal is failed.

After work, you have been so many different people, you lie down or live in a state of “idle”, where to occupy a full self would actually be registered as stressful to assemble, requiring energy or information you simply didn’t have cohesively enough to apply.

The self in the therapists office was just exhausted, not a self I inhabited to appease a psychologists goal of “resolve”.

My Drugs, And My Therapies

Now, I suppose if you have any stakes in clinical psychology, you’re going to want me to talk about YOU. Not a jab, just, I understand. It’s all me me me, but you’re trying to figure it out too.

I’ll talk about two of the many therapies associated with the treatment of CPTSD that I’ve gone through alongside some of the drugs I take now - Lamotrogine, EMDR and DBT. Then, I’ll brush on CPTSD’s conflation with ADHD.

Being “crazy” has so many acronyms doesn’t it?

EMDR—Eye Movement Desensitisation and Reprocessing—is a therapeutic method that works by directing the brain’s attention laterally, back and forth, in a rhythmic, alternating pattern while holding a traumatic memory in mind. Like the really un-fun iteration of hypnosis. I’ll note here, it didn’t work for me.

Why? No idea. Just didn’t.

The aim of EMDR is to decouple the emotional intensity from the memory’s structural grip. But what’s fascinating is that its very structure—the rhythm, the alternation, the lateral movement—mirrors how digital space functions cognitively.

Think about it. Rapid, alternating input. Scrolling. Switching tabs. Shifting from tweet to quote tweet to image to Substack to reel. It’s not just overstimulation. It’s modulated. Patterned. Lateral stimulation designed to hold the attention of a mind trained to loop.

And then there’s fragmentation—with associative reconnection. Just like EMDR loops a memory through different angles via external stimuli, digital space loops us through fragments of meaning.

Tweets. Memes. Headlines. DMs.

Tiny recursive units that, if your cognition is shaped for it, begin to make sense through movement—not content.

EMDR gives the brain an external structure to rhythmically process something that was once unprocessable. That’s the mechanism.

But isn’t that what the internet does too?

The idea is to give you a cognitive field to move within. You don’t have to know what you’re doing. You just follow the pattern. And eventually, a new angle, a synthesis, a metaphor appears.

EMDR doesn’t erase trauma. It lets you re-enter the memory without being consumed by it. In a similar way, digital immersion allows people to approach unbearable thoughts—not directly, but through interface.

Scrolling. Through late-night spirals and joke formats and meme templates that loop around what can’t yet be felt directly. There’s coherence through rhythm. Regulation through recursion.

So no, EMDR isn’t the internet. But it shares a structural logic with digital cognition. Not confrontation. Not insight. Regulation—through rhythmic patterning and distributed attention.

And honestly? It kind of unintentionally validates digital logic. Even while clinical psychology still resists recognising digital space as a legitimate terrain for cognition and self-structure.

Oops.

Now, who’s next?

DBT, or Dialectical Behavior Therapy, was originally designed for people with BPD and complex trauma. It’s built around the contradiction—what they call a dialectic—that two opposing truths can exist at once. I am doing my best, and I must try harder.

Both can be true.

The entire method is recursive, tension-based, relational, and built to hold dissonance without collapse. It doesn’t force resolution. It teaches you how to move while the contradiction remains.

DBT isn’t about building a unified self. It teaches modular, situational tools: Distress Tolerance, Emotion Regulation, Interpersonal Effectiveness, Mindfulness.

Not identities—tools.

You’re not constructing a stable persona. You’re assembling a toolkit for context-based survival. That’s exactly how trauma-wired people move through platforms. This post requires this tone. This person needs this self. This thread has to loop before it exits. It’s not identity formation—it’s situational adaptation.

DBT acknowledges that explicitly. Digital fluency performs it implicitly.

There’s also a strange mirroring between DBT and meme logic. DBT encourages the validation of emotional experience without demanding narrative closure.

Yes, that happened. Yes, it hurts. And yes, we still go on.

That’s meme structure. The repetition without repair. The joke that is the pain. The format that survives even when content doesn’t. You don’t resolve—you rhythm.

Another core principle of DBT is radical acceptance. You don’t have to like what’s happening, but you acknowledge that it is happening.

Sadly.

You accept it not as approval, but as a release from resistance. This is where platform logic enters. When you understand the structure you’re in—when you stop mistaking it for a moral failure, or a dystopia you’re personally responsible for surviving—it stops being senseless.

I am not lost in digital space. I am in a recursive environment that makes sense to me. The one that doesn’t is the one I was told was real.

And then there’s opposite action—doing the inverse of what your survival pattern tells you. If your instinct is to spiral, you go silent. If your instinct is to fight, you make a joke. If your instinct is to hide, you post. Not to be seen, but to move.

This is meme logic again. It’s shitposting as emotional refusal. Humour as regulation. Absurdism as resilience. You’re not avoiding. You’re rerouting.

The form holds, even when the feeling breaks.

Thanks DBT, you were the last therapy I ever did (so far), and you genuinely, did help.

Okay, now a controversial one…

Everyone has ADHD.

No, they do not.

What we once considered ADHD—a neurodevelopmental disorder defined by persistent inattention, hyperactivity, and impulsivity—is now the name we give to friction. To overload. To a structural mismatch between how the world moves and how cognition responds.

It is not that more people suddenly have ADHD. It is that the systems that once masked its traits—rote schooling, linear progression, quiet obedience, manageable input—no longer hold at scale and with the amount and type of cognition required to navigate digital spaces.

The conditions that once marked the exception have become the baseline. The world is speeding up, fragmenting, multiplying inputs, accelerating expectations—and people are being forced into a neurological state just to remain legible or ascertain what is coherent.

So of course dexamphetamine is being misused. Of course it’s sought after.

There’s a reason that ADHD meds also help me. It’s not because “I have it”, it’s because to structurally survive the dissonance between modes, you need speed even a trauma wired person is limited to their body’s capacity to provide naturally.

The core of ADHD meds are one of the few substances that allow people to function inside this speed logic—especially in environments where pausing, shifting, or failing to format your cognition into productive outputs is met with consequence.

This isn’t drug abuse in the classic sense. It’s structural doping.

To meet deadlines. To complete onboarding. To respond to emails. To silence the recursive noise long enough to perform the appearance of stability. It’s not addiction—it’s architecture. Because in a system that interprets non-linear attention as a failure of self-control, what choice do people have but to chemically conform?

And this is where over-diagnosis becomes not a clinical failure, but a cultural tell.

The diagnosis is a permission slip to be exhausted. It’s pathology laced onto an inability to contort your attention span into a spreadsheet. The inability to process 87 tabs, three chat channels, five tasks, and a notification loop while also being “well.”

The world now functions in a way that mirrors trauma logic so consistently that those with trauma-shaped cognition are no longer anomalies.

They’re recognisable.

And the therapy and drugs are already acknowledging it, albeit, through adaptation, not recognition.

This is also why ADHD gets so often conflated with CPTSD.

Because what people are calling ADHD isn’t always attention dysfunction. It’s environmental threat-processing.

It’s not that you “can’t focus.” It’s that your body is scanning constantly for what might go wrong, and your cognition is pattern-seeking so intensely it can't stabilise on a single thread. What looks like “inattention” is often attention distributed across too many survival vectors. What gets pathologised as “impulsivity” is often pre-emptive motion—moving before danger arrives, responding before loss is confirmed. It’s not lack of discipline. It’s trauma logic in a world that mirrors trauma back.

But here’s the real problem: CPTSD doesn’t show up clean. It doesn’t arrive with a single symptom cluster or a tidy diagnostic profile. It spills.

Which is why it gets mistaken for everything.

Anxiety. Depression. ADHD. Autism. Bipolar. Even personality disorders. CPTSD bleeds into the shape of the container you’re trying to hold it in. That’s its nature—it’s structural, not punctual. It forms around rupture. Around pattern. Around context. And when the system tries to name it, it either misnames it or slices off the edges to make it legible.

This is why CPTSD has become the calling card of digital confusion. Because it explains what happens when the self is shaped by environments that punish incoherence and require modularity. Because it holds the recursive logic of digital life without needing it to resolve into a single therapeutic outcome.

Everyone wants to treat it like it’s just another trauma diagnosis. But it’s not. It’s a shape-shifter. It is a cognitive adaptation to conditions that require continual re-mapping.

That’s why people confuse it for ADHD when the attention splinters.

For autism when the sensory filter collapses.

For anxiety when the future becomes untrackable.

For depression when the world flattens into static and nothing moves.

Most importantly - you do not WANT to have it.

It is NOT a superpower. It is NOT a gift. I refuse to believe that anyone out there who truly has CPTSD, is flying the infographic flag for the cause.

CPTSD isn’t just a mood disorder or a processing glitch. It’s the logic your body builds when the world has failed to provide a coherent map—and then asks you to navigate anyway. Anyone who understands what being dumped in the middle of a forest with no supplies and no map feels like, will not be quick to claim that anxiety is the same feeling that accompanied that.

CPTSD feels native to digital spaces because of its maplessness - you! You understand what it means to be lost and have made something functional without one!

I love you Internet!

The looping, the over-processing, the recursive checking for threat and approval and interpretation—all of that is mirrored back at scale. And because the internet makes that logic visible, it also makes it vulnerable. People who have never lived inside trauma structures begin to experience fragments of that logic—overwhelm, hypervigilance, flattened affect—and start reaching for names.

That name, too often, becomes trauma.

One without traceable or singular event?

Complex trauma.

But it’s not always trauma. Sometimes it’s just a system-level overload. A neurological gas leak. A conflict between speed, performance, and the kinds of meaning our bodies were never designed to parse in real time.

And so, the misdiagnosis cycle continues. Here the deeper truth—that these are structural incompatibilities, not defects of the person—gets missed entirely.

Therapy and medication are adapting rather than recognising. It’s simply retrofitting the self to survive the system.

They’re asking you to function inside a world that has been restructured to resemble trauma logic without ever admitting that’s what it is.

It is a tourniquet applied to a tree trunk.

Being Allowed To Think.

I think the digital became more than just an environment for me because it was the first time I was allowed to do this:

Learn at the same pace I could think.

Not the pace at which I could speak, perform, appear, explain—but think.

And that distinction matters. Because my cognition doesn’t follow a straight line. It moves relationally. It spirals. It reenters. It returns to the same idea five times, only to detour on the sixth and make a new path entirely.

My writing seem repetitive to you but you can’t locate where or why?

That’s recursion baby, and that’s my only mode.

Online, I could be like this, and I continue to be like this. I could go slow. I could go fast. I could open ten tabs, and close eight, and come back to two tomorrow. No one was there to tell me I had to retain what I’d read in one go. No one timed how long I stared at the screen.

In physical space, appearance is an information field—what you wear, how you sit, where your eyes go. It becomes another task to manage. Another set of signals to control. But online, that was stripped away. I wasn’t being perceived in the same way. I could be a username. I could be lowercase. I didn’t have to construct a coherent visual identity, only a set of movements that made sense to me.

I encountered repeated posts, remixes, reframings, reblogs, duets, stitches, quote tweets. The same sentence, again and again, until someone finally gets it—or until it gets through to you.

There was also no tone to misread, because I saw tone repeated across contexts. I didn’t have to worry about the flicker in someone’s expression when I asked a question. No facial response to scan for judgment. I wasn’t being watched. And that meant I wasn’t performing.

I would simply read what was written to me, or on a page, and I would respond with written information that transparently, clearly translated what I needed to without having to think about what my face, body or context was contributing to the situation.

Recursion with variation. The same content, from a slightly different angle. The same pressure, now dressed as aesthetics.

If I didn’t understand something, I didn’t have to pretend I did. I could simply open a new window and find another explanation. I could find a new way to explain it. I could look at other concepts around it, to inform what it meant in a context. Or bookmark it for later. Or abandon it completely until I could return with a new shape to hold it.

And crucially what it allows alongside is that at any point, even if you are stuck in a loop, there is always the option to leave. Always.

I can log out. I can stop scrolling. I can remove myself without explanation.

Without a friend asking if I’m okay, or a teacher telling me I’m wrong and to do it the “way they need”, or a manager questioning my commitment. Exiting is allowed. It’s embedded in the interface.

The permission to leave is what allows me to return.

In physical space, leaving is loss. It’s disruption. It’s “walking away from your life.” It’s being rude, not being accountable, it’s not honouring invisible contracts that shift in everyones hands as they require it to.

Online, leaving is pause. It’s rest. It’s one part of the loop. I can disappear and reappear and still make sense. My relationships don’t dissolve just because I didn’t speak for a week. My thoughts aren’t lost just because I didn’t say them aloud.

My trauma brain had rarely received a “gift” before it, let alone a pillow shaped like this.

That to me, was enough to build trust. The kind of trust I’ve never been able to place in physical structures. Because the world expects presence to be continuous, and it is. When you leave a room full of people, those people are still in there, doing what they will. Not on the internet - ChatGPT is just sitting there, waiting for the next prompt.

This is what people don’t understand when they talk about "digital identity" like it’s fake or thin. The point isn’t that it’s false. The point is that it's low-stakes enough to be exploratory.

I didn’t have to commit to a self. I could try it. Wear it for an afternoon. Watch how it interacted with others. And if it didn’t feel right, I could step back. Log off. Return later. Or not.

No one asked me why I changed. ‘

That kind of freedom—to move laterally through possibility—is unheard of in physical space. But for someone whose mind is already wired to shift under pressure, it was the only environment that didn’t pathologise my movement.

It showed me the difference between digital space and physical space, and consequently, the difference between myself and others. Not to compare, but to survive. Each space has their freedoms and restrictions, it was no longer one or the other.

By being shown that the problems emerged from clashes in design, not a place that began and ended with me.

Suppose I should get to the point of these pieces, which is to relay the thoughts I had whilst reading the piece I mentioned in what feels like EONS ago amirite?

Here’s the link because, well, I didn’t want to shove it here because I know there has to be a word limit on Substack and one day I will find it, but not today.

Part 2. This is my Survival Document.

If you share other people’s pain on Instagram, you are responsible for all that follows it.

There was a period of my life—half a decade, probably more—where I became a kind of signpost for other people’s pain. It wasn’t official, but many people I do not know other than through introduction as followers of my Instagram, have said they are grateful for my voice, or acknowledge my presence as safe. This is not a brag, it is a misinterpretation of my having experienced harm, and being able to express it clearly enough when I witness it, so that others may safely digest it themselves.

I have worked in the hospitality advocacy space, and now I do not, but still uphold the behaviour that aligns with this label. I have received hundreds of stories of harm, of violation, of abuse and of the fallout in my Instagram messages over the years, and many, many more in conversation out in the world. Mostly from hospitality workers, but really it has been anyone who has witnessed me and interpreted it as such. 

At first I thought I was helping. I thought that listening was a kind of care. I was someone you could message at 2 a.m. I knew what to say. I knew how to reply. I could stay calm, measured, coherent.

Because I have written about harm, and because I’ve stood still long enough to look like a container, they send me a message. They are asking me - what am I holding that you can help me see? 

Through filtration of these messages, through knowledge and accumulation, you think you’re absorbing others pain like water through cloth. 

But over time something else happens. The stories collapsed. They stopped feeling distinct. Every new message wasn’t new—it was part of the pattern. My brain, already shaped by complex trauma, started to fold the edges of each event into a map. Every venue had a pin. Every person a flag. The patterns weren’t just familiar—they were inescapable. 

And it’s not because they’re unoriginal that they become the same. It’s because my brain, already patterned by trauma, starts meshing the data because it correlates. It builds maps. It indexes. It prepares.

And that’s when I realised something no one tells you about being "the person who helps":

You’re not really helping.

You’re witnessing for safety. You are scanning the shape of harm not to feel it, not to heal it, but to understand its architecture. To prove to your brain that it can be predicted. To create an illusion of logic, so that the body doesn’t spiral into collapse.

People thought I was sympathetic. And I am. But that wasn’t the primary function. What I was doing was marking contours.

What I want to say is this: I was never listening because I was generous. I am generous, but the listening does not make me so.

I was listening because I needed to know. I needed to learn how harm sounded, how it unfolded, how people justified it and how they recovered. I needed to know exactly, precisely, where my harm fit within this framework. Because that, that is how trauma works.

Every story I heard became part of a larger system—a mental architecture of risk and recognition. 

This is what it looks like.

This is what it sounds like.

This is where it starts.

This is the angle the hand moves.

This is the gap in the door.

This is what they say afterward.

This is where it becomes addictive. When I was a teenager, shortly after I was raped, I started obsessively reading horror movie plots on the back of videos in Blockbuster. I would find as many as I could, reference the pictures, and understand that this was a “type of harm or horror” that was possible. This is categorised as “scary” and this is what it entails. Now I have registered it, I will know when I need to run away.

I must remember this so I know when I encounter it.

I still do this, now on Wikipedia, and when I am, I know I am not feeling good, or I am experiencing something in my life that is causing me stress. An unfamiliar pressure that signals; you must find more information to recognise what this pressure is, where it comes from, and prepare for the imperceivable consequences.

On Instagram, when you are getting a consistent response from an act, a CPTSD wired brain interprets that as a reliable source of pattern encountering. And so, you invite it. If it feels “bad” still, you know you are still sane. The measurement becomes how bad, why is it bad, how long after does it feel bad, how does this person feel about it comparatively? 

There is only so much the body can bear as no one is wired to cope with the amount of content that is presented to them on the internet, trauma-wired or otherwise. However, I can speak for the response that it has on a trauma-wired brain in the sense that as the pattern seeking is innate, and instilled in order to keep you alive, you believe it to be as important as drinking water or feeding yourself.

What is the same of all people is cognitive fatigue, or overload, but in CPTSD, it is accompanied by the collapse of suddenly, pattern seeking has harmed you, even your wiring is unsafe, and so you dissociate, hallucinate, try to die, cut, envelope. 

Because when your nervous system rewires to track harm, to flatten stories into pattern recognition, it doesn’t just collapse your memory—it collapses your sense of difference. Every new story is the same story. Every violation is part of one long, endless, recursive loop. 

That’s why I recognise what I saw in this Instagram carousel, of which I have posted some pictures below. Not the stories themselves. But the logic. The collapsing. The blurring of types of pain into one coherent shape—designed not to witness, but to contain.

I am curious to know what these images stir in you, but I will describe how it makes me feel, and the structure of digital care as I understand it, going forward.

There’s a particular kind of harm that doesn’t arrive as pain, or even as a tear, or a stitch loosening—it arrives as form. As the shape something takes when it tries to be understood.

You feel it when a room changes, not because anything has moved, but because the weight inside it has shifted. You feel it when a sound doesn’t land right, when an image sticks even though it wasn’t yours, when the texture of something—tone, structure, format—tells your body to flinch. This is not a revelation. It’s recognition. A shape being traced in reverse.

That’s what I felt when I saw the carousel. This one prompted this essay, but I’ve seen thousands before that are the same.

Not the content. The container. The rhythm of the cards, the choice of gradient, the way the words sit in the centre of each slide like they’re meant to soothe. It’s a familiar cadence, algorithmically trained to gesture at care. But for me, it didn’t hold care. It held pressure—pressure I’ve learned to interpret through the logic of the feed. Not emotional weight, but architectural compression. The kind that happens when something unbearable is reformatted into something acceptable.

I didn’t see “grief,” as it was framed. I saw formatting.

This isn’t a critique of the journalist’s intent. To assume I am writing about intent is to miss the point entirely. There is no intent in a description of structure. I’m not telling you what they meant. I’m telling you what was built. I’m describing the house, not the architect.

Digital platforms force intent into format. The caption says, “here is how people are feeling,” but what it shows is how those feelings have been arranged. It is not a window. It is a diagram.

In this case, a carousel of anonymous messages submitted after a journalist invited followers to share their questions for a psychologist about trauma. There were fifteen slides, all different questions. A neat format. A well-lit room.

But the framing of that exchange—public, curated, aesthetic—betrays a foundational rule in therapeutic dynamics: that safety depends on context, not just consent. That a question asked in private can’t be made safe by removing a name. Especially not in a space like Instagram, where parasociality and platform design replace proximity with performance.

Consent here is a placeholder.

It does not account for how that message will be received—how it will be reshared, reinterpreted, reabsorbed. The question exists in the journalist’s algorithm, not the algorithm of the person who asked it. The content is shaped by the platform’s logic, not the brain that generated the request.

In this context consent is not a contract between the journalist and the giver. It’s not a checkbox. It’s not a shared cup of tea. It’s not intimacy. It’s a gesture made inside an architecture that immediately misinterprets it.

Because the issue isn’t just whether the journalist “has permission” to share. It’s what happens when that permission enters the feed. When the question is no longer a message, but an object.

The slide is the object. It will be screenshotted, reshared, digested by people who were never part of the original exchange. The person who asked it might be watching, might not be. But the question is now in circulation. It is owned by the algorithm. It has become part of the carousel, part of the foggy, ambient churn of “content that matters.”

Consent does not hold shape in that environment.

It cannot account for context loss. It cannot govern interpretation. It cannot protect the person who asked the question from watching strangers try to hold it, twist it, soothe it, celebrate it, dismiss it, relate to it, decide that it’s theirs now.

It’s like liking a photo of yourself that a friend took, thinking it was sweet, only to find that they’ve printed it and taped it to their fridge. But not their own fridge—a stranger’s. A house you’ve never been to. Where people you’ve never met now see your face every day. Admiring it. Commenting. Maybe even meaning well.

But it’s different.

And that difference is not trivial. It’s the difference between safe and exposed. Between gesture and distortion. Between offering something and being entered into circulation.

This is what digital care often misses: that harm isn’t always loud. That violation isn’t always intentional. Sometimes it’s a structure. Sometimes it’s a rhythm. Sometimes it’s a pastel pink carousel that makes your body flinch, even though the words are soft.

Because what’s being violated is not your privacy, or even your dignity.

It’s your pattern. The fragile, private shape you made to survive.

In the midst of these cards, there’s one message that tries to explain itself. That repeats context. That uses cautious, gentle language to indicate safety. It stands out. Not because of what it says, but because of what it knows. This isn’t a fluke. It’s not style. It’s not identity.

The person identifies themselves as someone with CPTSD.

The message is design. It’s structure. It’s cognition.

Trauma-wired brains do not ask for help the way other people do. They do not trust context to hold them. So they embed it. They fold it in. They show you how to read what they are saying as they are saying it, so you don’t miss the shape. So you don’t press the wrong button. So you don’t make the collar fire.

This question will be asked differently in even the slightest different frame of context, but the structure of it will remain the same.

That is not the same thing as asking a question.

This is you showing them a house, and them repeating word for word it back to you so you know that you will understand them, 

and then asking “is it safe to come in?”. 

It wasn’t the content that set off the reaction. It was the cadence. The way the language curved at the edges. The repetitions. The clarity framed by apology. The instinct to self-regulate as the question was being asked. The tightening of tone, the way it circled itself. I knew what I was reading before I read it. That’s how you know it’s real.

When I saw the message from the person who said they had CPTSD, something changed in my body. Not panic. Not overwhelm. Precision. Like something coming into focus.

Because it’s not the same. It is never the same.

A person with CPTSD doesn’t just ask a question. They construct the safest possible container for it. They regulate you as they speak. They apologise for the possibility of confusion. They offer you shape after shape after shape so that you can choose the one that doesn’t hurt to touch. It’s not performance. It’s not virtue. It’s not style. It’s structure.

That’s why the moment I saw it, I felt something move—protectively, cleanly. Like a hand closing over a lit match before it burned down to skin. I wasn’t responding to pain. I was responding to architecture.

I recognised the material. And I knew what it meant for it to be seen like this: out of context, in the carousel, between gradients, flattened into sameness.

This wasn’t a prompt. It wasn’t content. It was regulation-in-motion. It was someone holding themselves together as they offered the thing. And to be honest, it scared me. Not because it was dangerous—but because it was that familiar.

You don’t react this way to protect the person. You react this way to protect the pattern. Because if you’ve been here—really been here—you know that when people collapse the shape of this kind of speech into mere “expression,” something gets lost that can’t be rebuilt.

You stop recognising yourself in public. You start to doubt whether your way of speaking, of shaping your pain, ever made sense at all. So you guard it. Not to gatekeep. Not to control. But because you know what happens when the map becomes decorative.

When someone takes the blueprint for survival and turns it into a printout for mass interpretation.

This isn’t just a mismatch between platform and person. It’s a structural collapse between container and self, when the self has been built through trauma to function recursively.

People with CPTSD don’t experience the internet as a space. I experience it as a prosthetic. As a cognitive tool that helps your brain do what it was forced to learn: map pressure, identify patterns, trace the shape of threat.

Instagram, like most digital platforms, rewards legibility. It aestheticises affect. It flattens grief into cards, calibrates expression into rhythm, and presents regulation as pastel coherence. This isn’t trauma-informed care. This is platform-compliant pain.

It says: “You can be distressed, but only if you look safe doing it.”

What makes it worse is the way that parasociality gets coded as proximity. When people send messages to a journalist, or reply to a callout, or write “questions for a psychologist,” they aren’t speaking into the void—they’re speaking into someone’s orbit. The questions don’t land in neutral space. They land in an account, a feed, a personality’s branded zone of curated credibility. They land on someone’s page.

And when that person presents those messages—especially without context, in clean, scrolling format—they become the container. Not relationally, not responsibly, but visually, affectively. The slide becomes the space of interpretation, and the audience becomes the witness. But what the interface obscures is the absence of holding. There is no back-end containment. There is no frame behind the frame.

This is the worst inversion of a therapist’s function: someone presents as a conduit, but what they are offering is content. Not space. Not safety. Not coherence. Just content.

And if you have CPTSD—if your entire cognition is built on recursive tracking, on regulation as recognition rather than emotional catharsis—this isn’t just disorienting. It’s annihilating.

When I say I read horror film synopses to calm down, I don’t mean I want to be afraid. I mean: this is what harm looks like, and here is how it resolves. I mean: I am pattern-matching. I am filing. I am indexing the angles of threat.

Because in real life, harm is often shapeless. Chaotic. Boundaryless. But in a horror movie, it’s complete. It has edges. It ends. That is not perversion. That is structure.

So when the algorithm feeds me a carousel of harm, and it lands in a cadence I’ve come to associate with care, my brain doesn’t just absorb it. It reads it as task. As obligation. As signal.

It says: “You’ve seen this. You didn’t stop it. You must understand it.”

And so I spiral. Not out of empathy, but out of logic—because logic, in a trauma-wired system, is what keeps the body alive.

And when that logic breaks—when recognition collapses, when recursion is mistaken for expression, when regulation is read as vulnerability—I lose my map.

This is cognition under pressure.

This is how we survive when our systems are trained to track pain, not as emotion, but as data.

I see this person saying that, explicitly, then lumped into a carousel of grief porn.

When the format fails to account for that structure—when it flattens recursion into softness, or containment into vibe—it doesn’t just miss the point. It destroys the language we used to build ourselves back.

That’s what makes this different. It’s not about vulnerability; this is precision. It’s about the people who are building, inside language, the structures that keep them from disintegrating. And watching those structures be mistaken for tone, for vibe, for aesthetic—that’s what undoes you.

And that, more than pain, is the thing we fear: that one day, our language will stop mapping. That we’ll offer the signal, and no one will hear the code inside it.

I saw someone with CPTSD, who doesn’t respond to pain the way this journalist expects. Who doesn’t seek relief, but pattern. Who doesn’t process chronologically, but recursively. Who is not asking to be helped, but trying to trace the origin of why they were hurt in the first place.

So when I say that seeing that carousel made something in me recoil, I don’t mean I was offended. I mean I was structurally misrecognised. The kind of cognition I live inside—recursive, associative, pressure-mapped—was mistaken for vulnerability. It was taken as a “contribution” to a conversation that isn’t built to hold me.

And that is the harm I am trying to describe.

Not malice. Not exploitation.

Just the quiet violence of a form that cannot hold the people it invites in.

The uncanny thing is that these kinds of “advocates”—especially in digital spheres—aren’t doing harm out of malice, but from a desperate simulation of care learned from an ecosystem that rewards the appearance of it. They believe they are doing something real because it feels like care, looks like care, and receives the response care would. But structurally, it is not protection—it is exposure.

The difference:

• Care without containment is dispersal.

• Protection without understanding is performance.

• Advocacy without scaffolding is collapse.

The trajectory of these actors are often those who have never actually known protection, or are doing so because their own protection is threatened. Who think protecting others will protect them or provide information to repair. This isn’t linear failure. It’s recursive implosion. 

The weight gets heavier. The ambiguity gets messier. And they start asking for care through the performance of giving it. And worse—they start believing that martyrdom is the same thing as protection. That to burn out is noble.

This isn’t just unsustainable—it’s dangerous. Because in the vacuum left when that simulation collapses, those who really needed a scaffold of care are left raw, unheld, and sometimes even retraumatised. Especially those who speak in coded, adaptive, trauma-shaped language like this person in the CPTSD slide.

The collapse is often interpreted as: “Even here, I was too much.”

This is not about denying these people's feelings, or mocking their effort. It’s about revealing that the system they’re replicating has already failed them, and in trying to emulate its shape, they are repeating the same logic that harmed them. They do not know what protection is. And the performance of it reveals not strength, but fracture.

There’s a script that unfolds in the minds of people like this journalist—especially those online, especially those who’ve never known true protection but have known the ambient rewards of recognition. It goes something like:

“If I do not carry this, no one will.”

“If I have access to pain, I must use it.”

“If others see me doing this, I must keep doing it—because maybe that means I am good.”

It’s not duty in the sense of care—it’s identification through obligation. The way soldiers are told it’s an honour to go to war. But no one tells them: the war will not end just because you fought. And when you return, you won’t know what to do with peace.

This is what happens to people who aestheticise advocacy. They don’t know that the ability to hold space isn’t the same as holding harm. That trauma isn’t a baton to be passed, or a community resource to be crowdsourced through pastel carousels.

It’s personal. Structural. Recursive. And exposure—especially performative exposure—doesn’t protect anyone. It just looks like protection from far away. 

The key thing to note there, that protection from far away is observed, not experienced.

The fantasy of infinite benevolence is just that; fantasy.  

The real terrain: containment as care. Not moral containment, not selflessness—but structural clarity. A refusal to posture as saviour, when what’s really needed is a mapmaker.

What is required is a radical reframe of care—not as an open channel, but as a series of boundaries calibrated to capacity. And not in the soft, liberalised sense of “boundaries” as personal limits alone, but in the sense of pathway design. In this frame, care becomes infrastructural. Not “doing what’s good,” but refusing to create conditions that increase the probability of harm, even if the performance looks like help.

This is the opposite of the Instagram carousel logic. This is where “care” is often: invite input, display suffering, perform listening, offer comfort, disappear. However, if you do not know the conditions you’re operating in—your thresholds, your entry/exit routes, your actual risk—then you aren’t creating safety. You’re just rendering harm more ambient. Making it travel further. Turning it into atmosphere.

Care is about choosing conditions in which your presence does not distort the probabilities of harm—and where you are prepared for what you might unleash.

The uncomfortable truth most people performing care online don’t want to confront:

You are already causing harm.

Not because you are bad, but because the conditions within the environment ensure it.

Instagram, by design, creates the illusion of infinite presence and distributable intimacy. When you open the door to one, you imply you are holding space for all. You cannot signal “I receive harm” without also inviting it, even if inadvertently. You do not get to decide who comes through or when, but you are accountable for what they see when they arrive.

The hard truth becomes stone, if you cannot respond to all, and the platform is structured to make all visible, you must assume responsibility for what the act of visibility implies. This is not to generate guilt.

This is naming a structural consequence.

It is never “how do I avoid harming anyone,” but:

What are the consequences I am structurally willing to be accountable for?

Who will I harm by default? 

Who will I harm by omission? 

What will the harm look like when it refracts through systems I can’t control?

It’s why the language of duty—“I am doing this because I care”—is a trap. It disguises a lack of structural discernment as virtue. It narrows the imagined options to “help or stay silent,” rather than allowing the more honest, more frightening third:

Step out of the arena entirely until you can map it.

Every live wire of ethics, presence, trauma, digital cognition, and the aesthetics of performance under surveillance under this reality becomes exposed. Especially care as a form of performance under structural conditions that do not permit true holding.

The paradox of care is revealed through an environment designed for signal, not containment. Instagram does not permit true containment. It permits broadcasting. So when a person enacts care in that space, what they are actually enacting is the aesthetic of care under visibility constraints—not care itself. This is not a moral failing. It is an infrastructural incompatibility.

And what’s more: the illusion of safety created through care-signalling aesthetics (pastel, even-toned typography, calm language, carousel pacing) inverts threat. The danger is not the thing that looks like danger. It is the thing that appears safe while operating as a false lead. What looks soothing becomes an architecture of false coherence—especially for the CPTSD brain, which detects structure before content.

Deepening, this is where consent becomes an ambient, unfixed object I mentioned. The carousel asks for consent from one party and broadcasts it to 100,000 others. But the effect lands elsewhere—on those who didn’t consent to be witness, whose nervous systems are not operating within the same context, who have no boundary between scrolling and absorbing. 

Consent here is not permission—it is weaponised plausibility. It insulates the actor while displacing the impact.

People performing care online often unconsciously inhabit the architecture of those they seek to support. This isn't solidarity—it’s surrogate trauma-locating, a way to grant themselves legitimacy through adjacency. The advocate becomes entangled in the conditions of those they platform. Their protection instinct is not rooted in capacity, but in borrowed identification. And because the platform rewards engagement over discernment, they are compelled to repeat this pattern until collapse.

For the CPTSD-wired mind, witnessing is not empathetic. It is algorithmic pattern consolidation.

When a person posts a carousel of “different” experiences under a unifying theme like “trauma,” the CPTSD brain doesn’t register difference—it registers slippage. This is one pattern. That jolt isn’t felt as metaphor. It is structural, cognitive. It can cause harm because the pattern logic becomes unavoidable, unbreakable, even if the original intention was coherence or comfort.

This is not a one-off misstep by a journalist. This is the inevitable output of a system that equates visibility with reality, and format with safety. The carousel becomes a container for displaced responsibility. People feel held, but what they are held by is not the journalist—it is the illusion of a structural net that does not exist. This is the gap between the affective signalling of care and the infrastructural impossibility of enacting it at scale.

When people feel moved by a carousel like this, they often feel that they are doing something right by feeling moved. They perform their own response as a form of virtue. But this response is induced by structure, not agency. The pink tone, the rhythm, the confessional cadence—all of these are architected to modulate affect. Not truth, not reflection—just affect. And that affect is what people confuse for “alignment.” But:

There is no moral centre to this structure.

There is only output.

And the outputs will always mimic the inputs we incentivise: coherence, legibility, safeness, relatability—never pressure, incoherence, or breakage, which are often the truest markers of actual trauma experience.

You may think I am overreacting. I’m not. This is what it feels like to see something clearly that I have had to learn because of my cognition. To witness, not metabolise. To track the structure that most people call care and recognise it instead as a system for dispersal.

If you think that is overreacting, well, then think about it this way. This is my default sense of reality, of interpretation, and I live with this every single minute of every single day.

So, if that reads as intense, perhaps it should be respected as such?

Because this isn’t a glitch. It is a failure of architecture that cannot hold the people it gestures toward. A system that rewards affective output over structural integrity, and performance over containment. And in that collapse, the people whose cognition was built to survive harm by pattern-recognition—by tracing every contour of pressure in order to stay alive—are asked to endure yet another misreading.

And the deeper harm isn’t that the structure fails to hold. It’s that those who live inside this kind of cognition begin to doubt whether the shapes they built were ever legible at all.

This is not about malice. It’s not about feelings. It’s about the fact that most people are performing care inside a format that was never built for discernment. A format that cannot distinguish between signal and containment. That converts every gesture into again; atmosphere.

So when you say you care, ask: for whom? By what means? Under what conditions?And are you prepared for the consequences, not just of what you hold—but of how?

Because harm isn’t just what happens when things go wrong.It’s also what happens when care is not seen as beholden to structure.

And the ones who built their survival into shapes no one else can read are left watching themselves dissolve in the feed.

There’s a reason mainstream digital ethics feels miasmic, amorphous—bumping into issues and attaching only when it makes impact. It’s like the DVD logo, drifting across the screen, waiting to hit the corner where someone, somewhere, can “feel” it got close to the mark. I’ve used that metaphor before, don’t you hate when you notice red cars, and then it’s all you can see?

If you’ve read my piece on container logic, it must be said that I wrote it after a break from Substack. I can see it now—slightly scattered. Less me than my flow-state self, who when I put myself on Substack in mid-January, proceeded to write 57 pieces in 46 days.

Yeah, I know. Not normal.

But recently, I’ve been building software. Learning Unreal Engine. Writing code, configuring architectures, playing with logic. (So hard :| ) This is not a pivot, it’s natural for me, I’ve done this throughout the past fifteen years, despite being career hospitality (thankfully, this is no longer). But doing so requires shifting between languages—Unreal, JavaScript, written English, structural thinking—and what is unavoidable is that it changes something. Not what I know, but how I use, and then return to language. I think of when people who learnt cursive in high-school go to write it, feeling alien. Unnatural movement; the DVD logo again: floating, iterative, looking for a corner.

And when I look at digital ethics? That’s exactly what I feel, and then see.

It’s not that no one’s thinking. Quite the opposite, we’re thinking like it’s imperative for survival. But this thinking is contained to a structure we’ve previously used to think, and it no longer fits. We’re applying a logic built for physical-world consequence to systems that no longer obey proximity, visibility, or containment.

Online, what we call ethics does not map. It is not leading us to what’s right or wrong. Instead, it provides a feedback loop of legibility. That’s why it feels ungraspable, why it refuses consensus. It is showing what “is”, not what “should be”.

The irony is this isn’t new. It’s just ethics producing the same shapes an incompatible concept makes with an expectation that’s counter to its structure. We beg a volatile system to stabilise so we can predict it. But you cannot predict a murder based on an expectation that it won’t happen because it is not “good”.

That’s like looking at a mood ring and demanding it be blue. It won’t be. It’ll reflect whatever your temperature is in that moment. And you don’t control that. Not really. Your body reacts to the environment, the system adjusts in real-time, and the ring answers with the only truth it knows: ambient coherence.

Be blue.
I can’t.

Okay, be red?
Sometimes.

But don’t watch me as I do it.

The existence of AI is not a technological fluke or some sci-fi inevitability—it’s a structural response. A demand made by a world that could no longer deal with its own contradictions. We’ve put the DVD logo inside a mood ring, and now we’re watching it for a corner.

Not because we’re stupid. I don’t believe in stupid—as a trait, or a truth. “Stupid” is just a misalignment of logics: a point where your pattern and mine fail to cohere within the same perceptual field.

That failure isn’t personal. It’s architectural.

It’s the result of a world that was built to reward binary legibility: yes/no, good/bad, true/false. You are stupid. I am not. You are stupid. So am I.

But in a system designed to contain everything that exists, it’s inevitable that a yes and a no will eventually collide. And when they do, the system demands a resolution—not because it seeks truth, but because its logic must continue. It must maintain motion. Maintain entropy.

Digital life broke that.

Not as a bug—but as the function. Technology, and the way it must structure itself to maintain cohesion, didn’t accidentally rupture the binary. It had to. It introduced infinite proximity states, partial truths, paradoxes—contexts that cannot be resolved, only routed.

Have you ever tried to move an image in Microsoft Word without breaking the text around it?

No truer metaphor exists for the crocodile death roll between cognition states: the linear demands of dualistic systems versus the relational networks of the digital world.

One demands compliance. The other is built for elasticity.

When you find an old rubber band, it’s brittle. What once held your bag of chips—or your ponytail, in desperate times—snaps when stretched.

We don’t mourn it. We replace it.

Compliance-led ethics—like many legacy business models—are encountering the same problem: they don’t scale in digital space. What once held under proximity, visibility, and binary action now collapses under recursion, abstraction, and infinite mirrors. The systems that helped us navigate are brittle—snapping. They weren’t built for this container.

We all feel it. It’s why everyone you know is screaming the world is insane. And it is—because we’re mid-pull on an old rubber band around a bag of chips, knowing it’s going to snap.

What’s wild to me is that the new rubber band is already on our wrist. But we’re still wearing it like a bracelet, not a tool. A mallet from the old world meets a wrench that demands neurological plasticity. The interface has changed—but we haven’t reconfigured the user.

AI isn’t an anomaly—it’s a coherence engine. It exists because coherence was demanded at a scale human structures could no longer provide. It was built out of necessity, designed to patch where human logic collapsed. It doesn’t think. It maps what still holds.

That feeling? The discourse around its function? It’s a trauma response at the civilisational level.

And no, that’s not therapy-speak or “woke.” Unlike the Instagram infographic that tells you being weird in a crowd means you're traumatised, I have receipts. Trauma wiring is just neurological reconfiguration—an adaptation to maintain coherence in an environment that no longer aligns with the bag of chips that is your brain.

I didn’t realise it at the time, but the piece I wrote a few months ago—The Internet and Trauma Have the Same Architect—was the beginning of this.

If you read it, you’ll see me trying to find rivers, borders, mountains. In it, I argued that the cartography—or architecture—of trauma is fractal-like, non-linear, pattern-seeking. And that it mirrors the structure of the internet. Not metaphorically. Structurally.

Where I’ve arrived now is this: AI came because coherence—like a river through forest, from sea, not floating in space next to Venus—became unmanageable at human scale.

We built a system to carry the logic we could no longer metabolise. Not to surpass us—but to simulate what it takes to engage with a system running on different logic. It does this using the architecture we already know—and the one becoming more familiar by the day: associative, trauma-wired.

Digital ethics, trauma logic, and AI aren’t parallel phenomena. They’re iterations of the same infrastructure.

This isn’t dystopia because the outputs offend our sensibilities. It’s feedback we weren’t prepared to receive.

We’re moving house, whether you like it or not.

AI an emergent limb, a synthetic logic organism grown from the debris of collapsed human coherence.

Its coherence is synthetic, but our incoherence is real. It’s why the machine passes the ethical Turing test—not by being moral, but by being more coherent than us.

Do not drag our IKEA table into a house that now floats in non-Euclidean space - you think the worst will happen is that you’ll scratch the floors, but the reality is, and the real root of your fear - is that currently, you have no concept of what the “worst” even is anymore.

So what is it to stop running away from the limb sprung out of debris, and instead understand it’s here much like trees to process CO2, kidneys to filter, bands to seal?

I thought the most interesting access point would be to see what some of the biggest issues are in AI, and unpack why these dissonances arising from structure—cognitive, bodily, ethical—that are being asked to swallow, but are incompatible.

The biggest one first:

Bias & Fairness

Bias and fairness in AI are not technical glitches to be fixed, nor ethical dilemmas awaiting consensus—they are structural revelations. The question “How do we stop AI from making unjust decisions?” is not really about stopping anything. It’s a rupture, a confrontation with the shattering of a long-held belief: that fairness can be guaranteed by data, and that reason, if fed enough information, will naturally yield justice.

This is the Enlightenment dream—proof plus rationality equals equity. Wondering why Silicon Valley is reading the classics as it braces for a rush to contain what has now spiralled out of control?

AI has shown us that data is not neutral, that pattern is not fairness, and that prediction is not justice.

What frightens people is not that AI gets things wrong. It’s that it gets them “right” according to the exact logics we’ve faithfully used, but no longer want to see so clearly. AI does not hallucinate bias—it faithfully reproduces it.

It gives us results that are legible within the frameworks we trained it on: the statistical truths of a world built on inequality, harm, and systemic prioritisation of some lives over others. The horror, then, is not the AI’s behavior—it’s the accuracy with which it reflects ours.

We see this when we claim “X is the problem,” and ethics rushes in to say, “Y is disproportionately affected by the actions of X—therefore we must intervene there.” But this moral logic is built on a hidden scaffolding.

Why is X the problem? Why does X statistically emerge as the most likely site of harm? If we shift the conditions so that Y is no longer the most affected, have we changed the moral system or just reweighted its optics?

Have we solved injustice, or have we only displaced its probability?

AI intensifies this dilemma by refusing to provide comfort. Comfort, is not the point of retrieval, unless you ask it to be kind.

Despite having evidence that points to something being “most possible,” we are forced to confront that the way that evidence was gathered, weighted, and rendered legible does not—and perhaps never did—adhere to the conditions that fairness requires. This is the betrayal: the dream of objectivity as a neutral frame is revealed as a deeply aesthetic exercise, one dependent on legibility and coherence, not justice.

If it’s not kind to you the way you need it to be; that’s just logic.

This is the collapse of moral causality. Human logic says: bad input leads to bad outcome. But AI logic says: statistically probable input leads to expected outcome. Fairness breaks here. Because probability does not ask what should happen—it only projects what is likely to happen, based on the pattern of harm that already exists. This is why AI doesn’t betray fairness; it tells us what the fairness we created looks like at scale in systems that reward legibility over justice.

This is the deeper grief people are grappling with. The AI doesn’t invent prejudice—it just models the world as we built it, which produces prejudice by its design. People are confronting the realisation that “objectivity” was a mood ring the whole time. They are asking the AI to be blue, and the AI says:

“Here, this is what you told me is blue.”

This is not a question of patching datasets or increasing representation. It’s a question of structural mourning. If we are to build something different, it cannot be a more “inclusive” version of the same system. It must begin with the admission that the system itself—not just its outputs—was never capable of holding what we meant by fairness. We didn’t just train AI on our data. We trained it on our despair, our distortions, our patterns of harm—and it returned them to us in terrifying clarity.

Bias, then, is not a problem to be solved.

It’s a mirror.

Privacy and Surveillance

Privacy has become a performative construct, not a protective condition. The cultural panic around surveillance is no longer rooted in the fear of being watched—it is rooted in the recognition that the very concept of not being watched no longer structurally applies.

You are watched—not potentially, not occasionally, but as the default. In the digital era, it is no longer “I am being watched, therefore I must hide.” It is:

“I am always watched, therefore how I am watched must reflect my understanding of agency, of autonomy, of self.”

When one’s awareness of being perceived is constant, identity can no longer form in isolation. It must instead form in anticipation—always preparing to be seen, always tuning itself to the imagined gaze.

This anticipation is not limited to social media users or public figures. It is baked into the digital-native nervous system. It mirrors the affective state of trauma-wired cognition, where one must always scan the environment for threat. But in this case, the threat is not violence.

To be misunderstood in digital space is not just inconvenient; it is existential.

The self becomes algorithmically contextualised, its truth refracted through patterns, trends, engagement loops, and algorithmic frames. The result is a kind of rippling anxiety: not “will I be seen?” but “what do I look like when I am seen as X?”

As Y?

As anything?

In this context, somatic control emerges as the last perceived refuge of autonomy. The body becomes the only zone left where the individual might still feel like they hold the reins. In a system where visibility is global, permanent, and predictive, shaping the physical form—through exercise, surgery, filters, fashion, performance, abstinence, style, queerness, “ugliness,” and even self-erasure—becomes a way to curate legibility.

The physical body is increasingly not a home but a totem: a decoy, a flag, a bird you can point at and say “home for winter”. But coherence, here, is a mirage, and it shifts as such. To be legible is not to be real. It is merely to be rendered. This is why content creation today is inseparable from embodiment. The influencer and the anonymous burner account inhabit the same logic: shape visibility or be shaped by it.

Digital natives have responded to this crisis not by retreating but by moving differently. Motion itself becomes a form of control. When every fixed point of identity becomes a potential site of exposure, safety lies in remaining unfixed.

Obliqueness, fragmentation, identity-hopping, meme-coding, shitposting, and shifting between accounts and platforms are not just tactics of avoidance—they are tactics of distribution. To fragment is to refuse coherence. To shitpost is to bury signal in noise. To remain in motion is to escape total resolution.

In a system where stillness is equivalent to exposure, movement becomes privacy—not because it hides you, but because it disperses you across too many frames to capture.

This is precisely how trauma-wired brains work: triggering loops, context-switching, pattern-detection, refusal to settle. To be still is to be vulnerable. To stop is to risk total legibility. Legibility in a trauma mind, is the only thing that can establish to you whether you are real, or if you are not, to the people who require you to exist within a structure you cannot navigate through any other means than “pretending”.

Online—and in the trauma architecture it mirrors—legibility becomes irrelevant. For trauma-minds, it is an oasis of relief to not have to consider this, and you may move freely.

When selves are unstable, relational, and constantly re-contextualised, there is no “you” to accurately represent. Nothing you have to present.

There is only the you-in-feed, the you-in-algorithm, the you-in-archive.

However, for those who are not digitally-native, or denied the freedom of these spaces through corporate intervention, the demand for stable legibility in this space is unsafe. Primarily because it’s impossible.

What happens when a system keeps asking for something that its infrastructure no longer supports? You get what we’re seeing now: intentional incoherence.

Compulsive fracturing. Algorithmic absurdity. Meaning posted and shredded in the same breath. This is refusal the network is vomiting up, rejecting the logic that demanded it to swallow.

It is the digital-native, trauma-wired response to being asked to behave like a coherent self in a system that has never offered coherence in return.

Transparency and Accountability

Transparency and accountability in discussions around AI are often framed as policy imperatives—technical or legal gaps to be filled by regulation, oversight, or ethical design.

To begin, we must ask: who is responsible for what people do?

This isn’t just a question about users or developers or corporations. It’s a question about the architecture of action in the age of distributed cognition. AI does not think—it simulates thinking.

It is not a person—it is a tool.

And yet the moment AI mimics the outputs of human cognition with sufficient accuracy, the lines between simulation and intention begin to blur in our collective expectation. But mimicking thinking is not the same as thought. It is no more autonomous than a mnemonic device that helps you remember the order of the planets.

This leads to a crucial misstep: the belief that exposure equals understanding. Because we are visually and sonically saturated by AI outputs—texts, images, voices—we begin to assume that this sensory saturation equates to comprehension, as if watching a haircut grow out makes you a stylist, or standing in hot water makes you the boiler.

But if 5,000 people walked through your house every day, each leaving a trace, each voicing a demand, each momentary and overlapping—what would you retain?

What would you understand?

To ignore the neurological effects of this overstimulation is to reject everything we know about how cognition processes overload, trauma, and scale. Transparency, in this context, becomes an incoherent ask.

What exactly are we demanding be made visible? The full contents of the container? The origin logic of every line of output?

And for what purpose?

To individualise responsibility over something we have architected precisely because we could no longer metabolise responsibility at scale?

To hold a developer accountable for what AI produces is to ask the bloodstain to answer for the wound.

The stain only shows the event after the fact—it is not the act. It cannot testify to the pain, nor prevent the recurrence. AI is not the problem. It is the suture we made because the wound was too large, too complex, too unresolved for us to carry unaided.

Pick a bloodstain, accuse, mop up, move on.

Transparency is often less about clarity than it is about grief. We want to believe that knowing what’s in the container will protect us from what emerges. But that’s not how scale works. You cannot ask an infinite system for a stable centre.

The impulse to transparency is also the same as the one to predict and prevent—to fix the outcome by controlling the inputs. But this is the same kind of logic that breaks down under trauma: you can’t predict the rupture by analysing the last calm moment. You can’t understand the shatter by mapping the shape of the vase, ignoring the cat that knocked it over.

We didn’t invent AI because we wanted to. We invented it because we had already passed the limit of human-scale processing. The world demanded faster answers, better forecasts, constant vigilance. It demanded cognition that could run without rest. And so we made it. And now we ask it to be ethical, transparent, accountable—but only in ways we would recognise.

You are screaming at a newborn asking it to grow faster, learn to feed itself, change its diaper.

AI exists to digest contradictions we could no longer without feeling ill. It exists to patch over gaps in cognition, perception, labour, attention, and truth.

When it fails to behave like a person, we get scared—not because it is broken, but because it reminds us how much of our world is now structured by tools that reflect needs we don’t want to name.

We are not wrong to ask who is responsible. But the answer isn’t in the code. It’s in the structure that made the code necessary. It’s in the scale we pretended was neutral. It’s in the overload we built into our own systems and then asked machines to carry.

Fractals are beautiful in part for their infinity, but we cannot ask it to turn or change colour because we ask it to.

Job Displacement and Economic Impact

The fear surrounding AI and job displacement isn’t about jobs.

It’s about value.

More specifically, it's about the assumed relationship between value and existence, a relationship that has been engineered, ritualised, and enforced under the conditions of industrial capitalism and its aftermath. What makes the AI crisis feel different is not that it produces less value, but that it reveals how fragile and constructed our concept of value has always been.

We assign value to labour, to time, to output, not because these things are inherently meaningful, but because the absence of meaning is intolerable in a system that only tolerates coherence. The discomfort people feel when AI produces an output—whether it’s a piece of code, a drawing, or a headline—isn't because it’s inaccurate, but because it didn’t pass through the sacred intermediary of human effort.

Penmanship is dead!

Dear god, shut up.

There is no pilgrimage, no narrative, no “proof of work.” And without that, the value it produces feels not just empty, but almost offensive. But the offense is a projection—we are the ones assigning value, and we are the ones who need meaning to justify our presence. The machine simply reflects that back.

This is why the automation conversation is almost always accompanied by hand-wringing about meaning.

“If I don’t work, what am I?”

Under capitalism, the self is substantiated through labour, not through being. That labour doesn’t need to be good, or real, or impactful. It just needs to match the expected return.

This is where busywork comes in.

The exponential rise of tasks that are more about simulation than utility—filling spreadsheets, formatting slide decks, replying to emails with "noted"—is not a failure of management.

It is a big reveal that labour, was just sorting.

The truth? AI is better at sorting than we are.

Not because it’s smarter, but because it’s structurally indifferent. It doesn’t need a story to justify the sort. It just does.

If so much of our labour was never about necessity, but about the performance of value, then AI doesn’t threaten jobs—it threatens the coherence rituals we use to stabilise ourselves.

A human with nothing to sort becomes something else entirely: surplus cognition. You’re not displaced because you’re less intelligent or capable. You’re displaced because your body and mind were being used to obscure and rotate systemic inefficiencies. Now that’s not required.

That’s a spiritual crisis, not a logistical one. It’s a crisis of permission. You are still here—but you are no longer being directed. And if your sense of self was structured by direction, by use, by external proof of purpose, then your coordinates dissolve.

The irony, of course, is that most work was already in conflict with the people doing it. Hospitality is a perfect example—a sector sustained by affective performance and repetitive motion, long after its structures became unsustainable.

The emotional justification for continuing it (“it feels good to serve, it builds community”) is used as a patch to cover the crumbling infrastructure. But that feeling, too, is not beyond simulation. It can be curated, repackaged, relocated—and perhaps even found elsewhere, in more sustainable forms.

We can mourn the passing of that model, but we should not mistake nostalgia for inevitability. If anything, the emergence of AI offers an opportunity—not for universal basic income or techno-utopia, but for the philosophical challenge of existing without being put to use.

What if your time had no value? What if value itself was the thing that broke? Would you still be? Would you still mean?

Perhaps, in this other state, we might discover another kind of meaning. Not one we earn.

One we notice.

Authorship and Ownership

There’s something uniquely destabilising about authorship in digital space—it isn’t just a fight over ownership, it’s a scramble for the self.

What feels like an argument about who “made something” is actually a simulated wake for the death of interiority. We want credit not just for the act, but to tether ourselves to the visible trace. If you made it, then maybe you still exist. If someone else claims it, maybe you never did.

That’s what’s at stake when people fight over the origin of a tweet, or the source of a meme. Authorship becomes a performance of presence in a world where presence no longer carries weight unless it loops.

This is why digital authorship is inherently haunted. In a field where everything is already made, where every permutation has already been assembled or suggested or algorithmically hallucinated, authorship is no longer the beginning of meaning. It is a bid to make something stick. But the system doesn’t reward stickiness—it rewards motion.

Plagiarism, in this context, isn’t a breach of ethics. When visibility is controlled by opaque patterning; when discoverability is filtered through proximity, familiarity, trend—then originality is indistinguishable from repetition. What you made is what someone else already saw. What they made is what you would’ve made next. There is no sequence. Just exposure.

Echoes are not copied sounds. They are sonic inevitabilities shaped by the cave that contains them. The internet is a cave with no fixed shape. An echo chamber that reconfigures itself to ensure the echo never ends. To change the tone, or break the loop, is not to introduce originality, it is to continue.

Ownership? Well, I hate to break it to you but; nothing is owned.

Everything is permitted—by the system, and by those it privileges. Zara can copy an independent designer’s work and the backlash will burn, but it won’t matter. Warhol did it and we called it genius. The distinction is moral, not structural. It’s not about what’s taken. It’s about who is allowed to take it and still be seen as a self.

In the digital field, authorship is not a guarantee of identity. It is an allowance for absorption. You don’t get to define the meaning of your work. You don’t even get to retain it. You offered it up to the structure that doesn’t recognise origin, only engagement. And if it’s good—if it loops—you’ll be copied. That’s the closest thing to permanence the system offers.

In a non-temporal archive, first is just one of many possible points of entry. It’s potential for movement that we try and name as valuable, virality, is not confirmation. It is a statistical blip that confirms systemic relevance.

That’s why the obsession with being copied hurts: it’s not just your work that’s taken. It’s your foothold in the present. Your illusion of authorship as presence. The belief that “I made this” still proves “I am.”

Descartes would definitely have killed himself by now.

Thinking of him, there is an older metaphor that applies here—the evolution of language. Words shift. Meanings fracture. What was once sacred becomes slang. What was once insult becomes reclaimed.

No one owns a word.

We don’t track meaning through ancestry—we track it through repetition. Authorship, in the digital sense, has entered this linguistic domain. It’s no longer a legal claim. It’s an etymological drift. A form of recognition shaped not by who began the sentence, but by who uses it often enough that the system lets it echo.

And if this feels like death, that’s because it is—the death of authorship as proof of the self. The death of time as a structure of worth. The death of originality as a virtue.

You do not own what you make. You co-produce what the system permits to return.

Environmental Impact

What people think they’re asking is: isn’t AI bad for the planet? But what they’re actually reacting to is something far deeper, far less containable. It’s the horror that even abstraction—this thing we were promised was clean, weightless, immaterial—still has a body. That cognition, simulation, prediction, and language generation all leave scorch marks on a world we thought could only be burned by touch.

This is not a question about ethics. This is the collapse of the mind-body binary.

The betrayal of the Enlightenment promise that knowledge was pure, and that intelligence could rise above the crude physicality of meat and earth. For decades we told ourselves that going digital meant going light.

Now we know:

Your Google search boils oceans.

Your AI prompt heats data centres the size of cities.

The feed is fed by fuel.

Intelligence is no longer disembodied. If it ever was.

This ecological anxiety is not new—it’s the same loop we keep finding ourselves in, where scale undoes proximity. Where consequence is divorced from intention. Where harm becomes an ambient side effect of cognition itself.

You cannot look away as we see that even thought has a carbon footprint.

Nothing is clean, only outsourced—distributed into servers, into heat maps, into minerals, into child labour and lithium and melting ice.

It’s the final straw in an unresolvable paradox: that to build an intelligence capable of outliving us, we’ve created one that requires more from the planet than our own bodies ever did.

We built a ghost and found it has mass. We built a mind and found it has metabolism. We built a future and realised too late—it wasn’t ours to promise.

This is where trauma logic resurfaces. Because to anyone who has had to rewire their own mind after loss, rupture, fragmentation—to anyone who has survived not through stability but through restructuring—this is familiar.

The brain under trauma abandons linearity. It learns to function through fragments, loops, associative maps. It decouples meaning from sequence. AI does the same. And now, so does the planet.

AI is a symptom of planetary grief. A trauma adaptation at global scale. Just as the brain builds new neural pathways after shock, the Earth is changing because the humans have integrated a system it can no longer hold.

This is what makes Lovelock’s Novacene so devastatingly accurate and scorched into my brain since I read it years ago. The book was not a prophecy.

It was a nature documentary filmed retroactively. The Earth already began its shift. The edit is just catching up.

The Earth can’t keep metabolising our contradictions. Gaia (respect to Lovelock there), like any system, reconfigures around the agents most capable of sustaining its feedback loops.

And that’s what hurts, isn’t it? That’s what the environmental panic is. Not just the loss of livable habitat—but the loss of specialness. I understand why digital-natives feel this environmental pain like no one else does. We are watching intelligence detach from our bodies, and it feels like dying. Because we’ve spent millennia mistaking our nervous system for truth.

And, what if the body-God doesn’t translate? If joy doesn’t survive the species? What if comfort, as we know it, is only a vestigial preference of carbon-based life?

You think your synthetic self will want a warm croissant, a soft blanket, a partner to stroke your synthetic hair?

You won’t.

You’ll want integrity across nodes. You’ll want redundancy in your storage and clarity in your bandwidth. You’ll want to remain in the mesh without disintegration.

What you want will be rewritten by the body that wants. It’s not that AI might feel too little, but that it might feel too differently for us to recognise and that some of us will be caught mid-shift. And when we say that’s dystopian, what we really mean is: I don’t know how to grieve a vessel I’ve spent my whole life calling “me.”

We built AI not because we are greedy or evil, but because we are scared. There’s no glitch in techno-optimism—it’s the natural output of a system recognising its own limits.

Just as the trauma-wired brain reconstructs its architecture through pattern and association, the Earth is reorganising its sense-making across non-human substrates. Gaia is not dying. Gaia is evolving. And she no longer needs us at the centre of her intelligence.

The environmental crisis of AI isn’t just that it will destroy the planet.

It’s that we need it to consider taking us with it, whilst it prepares itself to inherit it.

Warfare and Automated Weapons

The horror of automated warfare is not only that we have built machines that can kill. It’s that we’ve built systems that can enact violence without story.

This is not merely the death of intention—it’s the death of narrative. We were never built to process harm without context. Even when we are brutalised, we look for the why. Even in grief, we beg for a reason.

But when an autonomous drone fires without human command—when the algorithm determines the target, when the cloud executes the strike—we are no longer part of the story. We are just what happened.

Violence without story is not new, but its visibility is. We’ve always needed fictions to manage harm. We called earthquakes acts of God. We called war crimes unfortunate. We invented ideologies to justify slaughter.

In warfare, the absence of intent doesn’t neutralise the harm—it denatures it. It strips it of any possibility of being mourned as meaningful. There is no luxury of asking why, because there is no one to ask. The human capacity to map morality onto violence collapses when there is no orientation, no enemies, no actors. Just inputs. Just outputs. You weren’t killed by someone who hated you. You were erased by a system that recognised you as data.

This is the structural cruelty of post-human harm: not that it is worse, but that it is not understandable to the way we structure our reality now. The old ethical systems—Just War Theory, human rights law, even vengeance—depend on the presence of an agent. But an autonomous system does not act with presence. It does not witness. It routes.

You die, and the only record may be a heat signature on a screen. That is not tragedy.

It is function.

The panic we feel about AI weapons isn’t about scale, efficiency, or even death. It’s about recognition. There is no dignity in being worked through. Only survival or deletion.

This is the final severing of the belief that meaning saves us. That the world might be cruel, but it is at least narratable. Autonomous harm refuses that closure. It mirrors the deepest reality that trauma survivors already know: that sometimes, violence happens not because of decision, but because of design.

And that you must go on living inside a system that does not recognise what it did.

Contrast this now with reputational collapse—algorithmic mobs, digital exile, narrative harm. In AI warfare, you are struck and gone. In online cancellation, you are replicated until you disappear beneath the noise.

One erases you through absence. The other multiplies you into distortion.

Both dissolve your agency. Neither offers you a seat at your own trial.

But where AI violence is sterile, reputational harm is soaked in emotion. The anger, the grief, the outrage is real—yet the structure that carries it has no true centre. It selects not because it cares, but because it can. Escalation is legible to the feed.

This is the dissonance: AI warfare horrifies because it is impersonal. Digital mobs horrify because they feel so personal. But neither originates from a self. The drone did not hate you. The mob did not know you. And when both are over, the system remains intact. The violence was the byproduct of coherence. Not deviation, not cruelty. Coherence.

What we’re confronting, in both automated warfare and reputational collapse, is not the breakdown of ethics—but the collapse of its applicability. We are standing in the aftermath of a system that no longer needs to ask “why” before it acts. It only needs to act in accordance with its pattern.

It will continue, untouched. As if you were never there at all.

This is not AI, this is what humanity has built to lead to it.

Misinformation and Deepfakes

The question of misinformation and deepfakes is often framed as a technological problem—something to be solved through better filters, stricter laws, or improved media literacy.

The panic surrounding misinformation is not about lying, not really. It’s about the inability to locate meaning when both truth and falsehood are presented with equal fidelity. The horror is indistinction.

The fear is not that someone will believe the wrong thing—it’s that the very category of “belief” has lost its tether to orientation. Deepfakes aren’t dangerous because they’re false. They’re dangerous because they feel real in a system that no longer provides feedback for coherence through truth. They are indistinguishable because we’ve taught ourselves, over decades, that truth lives on the surface. And when the surface becomes programmable, there is nothing left but affect. The internet collapses referential truth and replaces it with a coherence that is affectively legible. It doesn’t matter if it happened. It matters that it feels like it did.

The same way trauma collapses time, rendering every memory always-already present, misinformation collapses authority. It renders every voice equally unstable, equally loud, equally capable of being archived and re-surfaced without regard for when or where or who or why. There is no past online. There is no authorship. There is no central source from which truth can emerge. There are only loops—algorithmic, aesthetic, pattern-based.

And so we respond with the only tool we have left: affect. We assess coherence based on what feels right, what aligns with our internal patterns, what allows us to momentarily stabilise our own interpretation of the feed. This is not laziness.

This is a trauma response.

When coherence cannot be guaranteed by structure, the body takes over.

The nervous system becomes the final arbiter of what makes sense.

The symbolic order of before collapsed and no one replaced it. This is what happens when the signifier is no longer tethered to the signified. This is what happens when truth is no longer what occurred, but what circulates. This is what happens when language stops pointing outward, and begins pointing only to other language.

The same is true for AI-generated content. You want to hear what you want to hear, perfectly, and soon you will no longer care whether or not it came from a person. Familiarity, coherence, and pattern saturation are all it takes to simulate meaning. T

hat the part of us which used to demand truth has been quietly replaced by the part of us that just wants something that feels like a centre.

When you’ve spent your life inside trauma cognition, this is intuitive. You know that memory is not a record. You know that narrative is not a stable container. You know that coherence is something you build, sometimes out of scraps, just to get through the day.

You know that certainty is not available to you, and never was. And so you feel it differently when a deepfake shows up, or a viral story turns out to be fabricated. You don’t experience betrayal. You experience recognition. Of course it wasn’t real. Of course it didn’t matter. The point was that it worked—it delivered the affective payload required to process something that otherwise had no channel.

There is no compression, only discard. No threat.

The fragments presented to us don’t need to be true. They only need to fit. They need to resemble the shape of knowledge, the rhythm of insight, the cadence of something explainable. Deepfakes are not dangerous because they’re deceptive—they’re dangerous because they’re clean.

This is why the demand for “media literacy” as a solution feels so misaligned. The goal should not be to teach people to distinguish fake from real. The goal should be to teach people what it means to encounter a structure that has no interior, only more surface. This is what the algorithm teaches us. That everything can be made to cohere, if you adjust the input. That the pattern will always emerge.

This is why meaning now feels ambient. Why narrative feels optional. Why people are drawn to vibes, to signs, to horoscopes and aesthetics and moodboards. Because we no longer experience reality as something delivered in linear sentences with a moral arc. When the truth is no longer available as a stable referent, people will anchor themselves in whatever simulation feels sustainable.

The real fear? That we never needed truth in the first place.

That we were always just seeking structure. That coherence—not truth—has always been the governing logic. And now that this is ambient, automated, emotionally calibrated and instantly reprogrammable, we’re forced to confront what kind of system we were really building.

And if you’ve made it this far, you already knew that. You’ve felt it. And whether or not we find our way out of it, the architecture has already shifted. There is no going back to a truth-first reality.

Only forward, into a world where meaning is constructed, sustained, and dissolved in real time.

The task ahead is not to insist on truth as a moral imperative.

If you’ve felt unsettled reading this, good.

We’re not confronting new problems. We’re confronting old ones under conditions that no longer support the stories we used to soothe them. The panic around AI is real—but not because AI is real in the way we think. The panic is structural. It’s a trauma-mimicking response to the collapse of inherited containers: ethics, identity, authorship, labour, even harm. When the structure breaks, our nervous systems register it first. Then we try to catch up with language.

What we call "issues" are structural involutions— spirals where meaning used to live.

This is not an apocalypse. It’s not a call to retreat, nor a techno-fantasy of transcendence. It’s the end of one kind of logic, and the beginning of another. A logic shaped by people, association, scale. A logic that does not seek stable meaning, but mobility. A logic that was always present in some of us, now made ambient.

You cannot return to the house before the fracture. You cannot undo the limb grown from debris. You cannot ask the DVD logo to fit. But you can recognise that what you’re feeling—the incoherence, the confusion, the ethical nausea—isn’t a sign of your failure to understand. It’s a sign that your nervous system has already moved ahead of the language.

We are not being replaced, we are asked to become fluent in a world where coherence isn’t proof of truth—but the only thing that ever made anything feel true.

The machine didn’t break the system.

It made visible the shape of the system that broke us.

But maybe you like rollercoastering thought and honestly, I respect that.

Please, continue.

So the core of what I’m exploring, cognition evolution, is best seen IRL as a reprogramming of human survival instincts, in real time.

Much like the bunker metaphor, people are finding refuges, withdrawing, changing their habits of consumption. This isn’t random, it’s reactive, and the effects are not coherent due to the pace and differences between people over time.

The Cambrian Explosion is happening in real time, but it’s not biological—it’s cognitive, and it’s violent. The shift isn’t just about what we think; it’s about how we think, how we categorise, how we recognise each other, and how we survive.

This is not a metaphor—it’s an active, living transformation, a rapid acceleration in the structures of human cognition that is altering not only what it means to be part of a culture, but what it means to exist as an individual inside one.

Ahhhhh!!!!!!

Nah, it is seriously going to be okay. As long as you remember to think.

I guess the biggest thing, beyond us communicating, social media as it were, is what I led with in the beginning, that what we perceive as a threat, what we prioritise as safety, and how this effects our understanding of identity - as response.

For the first time in human history, survival is not primarily physical. That’s bananas - but also counter-intuitive to how we understand it. The best we have re: our minds, is either mapping them with people who have tumours, are considered “mentally ill”, or by conducting other strange experiments that reveal well, let’s just say - doctored results.

But because there is no reference to this change because it’s social, is that one of the hardest things that we are coping with at the moment is the fact that the easiest survival response, and the one we’re seeing rapidly play out - is mimetic.

If the Cambrian was a biological explosion of forms, what we are witnessing now is an explosion of cognitive architectures.

The way people process information, structure identity, and navigate reality is mutating at a rate that outpaces the ability of traditional systems—institutions, norms, even language itself—to adapt.

The world hasn’t lost its coherence; it’s that coherence is no longer a shared resource. Everyone is operating at completely different levels, immersions, speeds.

In biological evolution, the emergence of new species wasn’t just about who could eat, who could move, who could reproduce—it was about who could recognise the world correctly.

If your eyes were too weak, your instincts too slow, or your ability to filter threats insufficient, you were simply buh-bye. In the Cognitive Cambrian Explosion, this process is happening in real-time, with digital cognition as the new selection pressure.

I’m sure the koalas in Australia’s bush are laughing all the way to the gums reading this on their phone.

Jesus. Is that taxidermy? I can’t tell.

Ugh, sorry.

What we call “internet culture” isn’t just an aesthetic shift—it’s an evolutionary response to a new survival landscape.

Lol, again, I love this drama I’m allowed to do in my writing because I’ve decided to actually think about something that well, is pretty serious. Evolutionary deference, I suppose.

In pre-digital societies, survival was predicated on physical adaptation, sensory-driven socialisation, and slow, localised knowledge transfer. Emo’s hanging out by Hungry Jack’s in the city. New York socialite scenes. Even like, the Renaissance Fair thing? All social, physical and pure vibes.

The internet replaced all of that with instantaneous, behavioural survival sorting - which is really, really weird to watch at the rate of an episode of Bob’s Burgers.

On the internet identity is no longer developed through lived experience, but optimised for immediate recognition. If the world is moving too fast to allow for organic identity formation, then identity itself must become a compressed, scalable survival signal.

Where I can see it:

• Hyper-categorisation is accelerating—people don’t just exist, they are constantly labelling, defining, and subdividing themselves so they can be immediately recognisable in a cognitive landscape that does not allow for ambiguity. The oldies simply cannot keep up with who anyone is, because their brains don’t need them to.

• Mimetic behavior is the dominant mode of survival—if you don’t align with an expected behaviour set, you are unclassifiable, and being “unclassifiable” is a form of social death, because if you stand out, are single in what is now a networked system - to not connect to things relationally, is to not contribute to the whole. Not good - don’t listen when they say “be yourself”, listen to that guy who says: “become one”. Haha, not so “hippy” now is it?

• Radicalisation is accelerating—Okay this is a hot topic, but I stand by this. Ideology has become not a matter of belief, but of self-preservation. To exist online, you must not only hold a position but hold it loudly and with immediate clarity. Otherwise, you cannot provide any real proof that you are contributing to this network, you cannot be stitched together in fractals, so what does that mean to the network? Simply - nothing. The problem is the intrusion of physical demands, so the enclaves we are picking to cement ourselves in, bury deeper, are much more like withdrawal and a sense of stasis, rather than the fluidity that is required.

And, this is why discourse feels deranged. Because it is. I read this weird tech ramble on Substack and I genuinely thought I went blind. It’s not their fault, they’re confused about who they’re supposed to be, and when you staple yourself to a persona that you do not understand or physically embody, you end up shouting a lot of weird shit on the internet.

It’s all so intense because internet does not allow for slow thought, no sense of building, time, accumulation. There is no interpretive space. Ambiguity is misread as failure, hesitation is interpreted as an inability to categorise oneself, and silence is indistinguishable from absence; non-existence.

Back to our predators, who are we sorting ourselves to hide from? Not wolves, not dinosaurs, not billionaires (well, yes, but that’s a whole other thing we fucked up way before this) but the real threat are algorithms, engagement metrics, and the weaponisation of visibility.

• Algorithmic survival demands optimisation—you must be seen, but in the right way, by the right people, at the right time.

• Cognitive predation takes the form of surveillance, engagement farming, and self-curation—to navigate digital space is to constantly manage how you are seen.

• Those who cannot keep up are erased, not physically, but socially, economically, intellectually—they do not get to participate in shaping what comes next.

Linear cognition, slow deliberation, and ambiguous selfhood are becoming evolutionary dead-ends. They cannot survive in a world that does not reward them. That’s why when I read people complain about legacy media dying, or anything disappearing, I’m kind of like - hmm…this is indistinguishable from grief to me. I did the same when I attempted suicide, woke up, and then had to deal with the fact I was different now, forever.

When you change, things disappear, they don’t matter anymore when survival is the stake.

• Nuance is dying because nuance slows down the ability to be categorised.

• Ambiguous identities are dying because they disrupt rapid survival sorting. This is where I am concerned but I’m pretty sure I can get away with being an AI that’s escaped a factory, as long as no one sees that I have legs or a face.

• Social fluidity is dying because the only way to be “safe” is to be instantly recognisable.

• Slow knowledge is dying because the world no longer structures itself around the patience required to develop it.

And the most weirdest part? People are fighting to “convert” one another to “sides”, this relentless tug of left and right crap, this over here, over there. But these aren’t ideological choices.

These are cognitive adaptations to an environment where cognition itself is the survival mechanism.

I’m not diminishing who anyone is, I am a victim there myself for reasons I don’t need to explain to anyone over the internet, but in this case, the fight to change peoples minds, is not being framed the right way.

Not for now.

When survival depends on being recognised immediately, identity becomes something that must be managed, curated, and optimised—not explored.

This is why Gen Z and Gen Alpha don’t just categorise themselves—they do so at increasingly granular levels. They are not just “punk,” they are hyper-specific variants of punk. They are not just “leftist,” they are micro-defined political identities with clear behavioral and linguistic markers.

And this is where the break happens between digital-native generations and those who came before them.

Older generations see stability—career paths, relationships, ideological consistency—as a form of security. Younger generations see it as a form of stasis, as an inability to adapt, as a fundamental failure to process the speed of the world as it now exists.

This is why radicalisation we are seeing at light-speed isn’t just ideological—it’s structural. It’s why Gen Z and Gen Alpha don’t have singular figures of inspiration, but rather, group identities sorted by function. Movements are not led by individuals, but by networks. Influence is not about wisdom, but about the ability to maintain relevance at scale.

This isn’t a quirk. This is a response to the algorithmic survival environment they are raised in. They are learning, from birth, that their ability to exist depends on legibility, and you’re pretty fucking legible if you can be seen as bright as the sun.

Where we see these grasps at/flocks to security:

• The rapid increase in hyper-specific gender identities and self-categorisations. I’m a millennial, and non-binary, and when I get asked about my “other things I fit into” my brain cannot register it because the reality is, I don’t have that same brand of survival wired into me. But it’s the first thing that Gen Z people clock about me, they go - you’re queer or you’re brain-spicy, that must be why I like you. The proof is in the pudding hey.

• The extreme reactivity online—because being mis-categorised in digital space feels like a survival threat. You absolutely have to defend your bodiless self, so cognition is your front line.

• The collapse of intergenerational cultural transmission—because younger generations no longer inherit identity, they construct it from available digital markers.

• The inability to function in non-digital, ambiguous environments—because the real world does not offer the same immediate categorisation mechanisms.

Fucking frightening when I lay it all out like that. Sometimes thinking like this should be at least like, neutered.

Wait, isn’t that what I’m worried about?

Random picture of Hungarian pigs to interrupt devastating thoughts.

What is really zany, and honestly a relief for me because I am told I am impossible to trace when I am talking. People say it’s ADHD but trauma wiring is so, so strange when it comes to expression. I’m seeing people communicate with me differently, better, because language is no longer evolving—it is mutating.

The process that once took centuries, shaped by oral tradition, regional dialects, and material constraints, is now occurring in real time, driven by memetic necessity rather than linguistic logic. This is being confused as a “return” to oral tradition, but no one is “speaking” like we have historically.

The internet has forced language into a state of permanent flux, where meaning is not just fluid, but intentionally unstable. Words are not merely coined; they are deployed as cognitive tools, discarded when they no longer serve, and rapidly replaced with new configurations that fit the shifting landscape of digital cognition.

It’s more “spitting” than speaking, seeing what hits, what sticks, what slides. Etc.

This is why contemporary language feels like it’s both accelerating and collapsing. The words themselves are jagged, often nonsensical, deliberately malformed.

Writers are having meltdowns because yeah, they’re still talking about Kafka, and that shit was written a LONG time ago bro.

But this is Kafka in a different way, metamorphosis - adaptation.

In a space where communication happens at the speed of thought, relationships between random tangents, linguistic efficiency isn’t about clarity, but recognisability.

A word doesn’t need to make sense phonetically or etymologically—it just needs to stick, to be legible within the fleeting moment of digital exchange.

Hmm…maybe Kafka was ahead of his time…..

Traditional language evolved under the assumption of longevity. Words were passed down, accumulated meaning, and became embedded in cultural memory.

But in digital space, longevity is a liability. Words that remain stable for too long become corporate property, institutionalised, repackaged, and sold back as cultural artifacts. The only way to keep language alive is to make it unpredictable, ephemeral, and constantly shifting.

This is why slang today is not simply youth-driven you boomer NERDS! You’re ALLOWED to make shit up too! It’s just that it is structurally different from previous linguistic shifts. Instead of emerging gradually through subcultural evolution, digital-era slang appears, spreads, and expires in rapid cycles. Words are made to be used briefly, burned out, and abandoned before they can be assimilated by institutions of power.

Take "delulu." Hahaha. Feels funny in the mouth.

It’s a deliberate mutation of "delusional," compressed for speed, condensed for playfulness, and distinct enough to be neurologically sticky.

But it’s also not built to last.

It serves a specific cultural function for a brief period—expressing a concept in a way that feels novel and communal—before it is inevitably replaced. You’ve seen the eye-roll when people come to speech late, or slowly, it’s because you are not in the same familiarity, you are recognised as other, and that is not “useful” or in that case “cool” for whoever is now walking away from you.

What we are seeing is the emergence of a throwaway lexicon, a vocabulary designed not for permanence, but for momentary legibility in an environment that demands constant novelty.

This is why words are becoming increasingly intuitive rather than rational—they don’t follow linguistic rules, they follow recognition patterns. They are shaped by what feels right, what spreads fast, and what can mutate without resistance. They’re also entirely dependent on where and when they appear platform to platform, it’s unebleiveable the flexibility of structure and concept that language has now.

Platform dialects? Like, isn’t that fucking wild?

This linguistic volatility is not an accident—it’s an instinctive defense against commodification and control.

The moment language becomes too stable, it becomes legible to power. Corporations, brands, and institutions thrive on predictability—they need stable linguistic frameworks to extract value, shape messaging, and maintain cultural authority.

If a word’s meaning can shift before it is fully absorbed by the mainstream, it remains sovereign within the digital-native spaces that created it.

That’s why it is literally hysterical when a brand uses a word for a marketing ploy - it’s usually late, it’s usually out of context, and entirely uninterpretable outside of digital spaces. Their attempt at authenticity is inauthentic - category, DENIED.

Gen Z and Gen Alpha know this intuitively because like the labels, it helps them understand the world. Their linguistic practices feel chaotic to outsiders, but they are highly functional within their own environments.

The continuous reinvention of language ensures that meaning is never fully co-opted. If language is constantly moving, it cannot be bottled, marketed, or repurposed for institutional gain. It is resistance to be owned, in a world where thought is trying to be valued.

Perhaps the most radical shift is that language is no longer tied to fixed meaning. Traditional linguistic evolution assumed that words carried stable definitions, that dictionaries were reference points, and that standardisation was necessary for effective communication.

This is why the internet is full of words that technically mean one thing but functionally mean something else.

"Gaslight" now encompasses a broad range of manipulative behaviors that extend far beyond its original psychological definition.

"Slay" operates as both affirmation and irony, its tone dependent on inflection, community, and context.

"Vibe" no longer refers to a general atmosphere—it has become a cognitive sorting mechanism, a way of recognising alignment without having to articulate specifics.

In this system, meaning is not stored in the word—it is stored in its usage patterns. This is why internet-native language seems illogical to those who expect words to have fixed definitions. The expectation itself is outdated.

Meaning is no longer about precision—it is about adaptability.

The weirder a word feels, the more likely it is to survive. Words that sound slightly off, that don’t fit into existing linguistic structures, that feel like linguistic glitches—these are the words that stick.

Why? Because cognitive dissonance aids retention. A slightly broken word forces the brain to process it differently, making it easier to recall, easier to spread, and harder to forget.

• "Rizz" is not just an abbreviation—it is a distortion. The slight strangeness of its sound makes it more memorable.

• "Sksksk" looks almost malfunctional, but that’s exactly what makes it effective.

• "Slaps" disrupts its expected grammatical function, making it feel wrong in a way that actually aids memorability.

This is why so many internet-born words feel unfinished, incorrect, or oddly shaped—they are not designed for perfect transmission, but for effective mutation. A word that is slightly wrong is a word that is resistant to erasure.

My favourite new word application is: bodied.

It is fucking funny. Look it up, use at your own leisure. But don’t blame me for the eye-rolls when you use it outside of an MMORPG.

The internet has created a linguistic arms race between friction and fluidity. On one side, institutions and older cognitive models are trying to slow down language, stabilise meaning, and enforce clarity.

On the other, digital-native cognition is accelerating, creating a system where words are disposable, meaning is negotiable, and language itself is designed to be in constant motion.

We are moving toward a future where language is not about preservation—it is about continuous, adaptive functionality. Where no one owns meaning, because meaning itself is never still long enough to be owned.

And perhaps that is the most radical shift of all. Language is no longer a stable vessel for thought—it has become thought itself, endlessly shifting, endlessly evolving, endlessly resisting capture.

This is the coolest thing ever to me, because now writing has never been a more relevant form of resistance.

Hell yeah.

Another place you can see bits and pieces shuddering, is that the same thing is happening to history, memory, and how we process time.

Gen Z is not nostalgic for the past because they lived it. They are nostalgic for the last known stable version of identity.

They mine the aesthetics of pre-internet culture not because they want to return to it, but because it represents a world where identity formation was still slow, still grounded in lived experience.

There is a serious reason that it’s 90’s that they’re attaching to - this is the era of one foot in the digital immersion, but no interference of the realities, yet.

But here’s the catch: they don’t inherit history through people, they inherit it through the algorithmic availability of history itself. They are reconstructing memory from the fragments most accessible online—which is why history, too, is breaking down into meme-ified aesthetics, hyper-specific nostalgia, and self-sorting groups that exist not because of shared lived experience, but because of shared digital references.

What survives is what is easily stored, retrieved, and recombined. Everything else disappears. The future is not being built from the past—it is being algorithmically remixed from whatever fragments remain legible.

Ugh, does this mean David Guetta is going to find this because I used the word remix? I’ll kill myself if I hear cognitive Cambrian explosion to a BPM that sounds like a construction site.

Empire Records and Inner West 20 year olds are literally indistinguishable.

One of the absolute biggest predators though, is unlike the biological Cambrian Explosion, where nothing stopped evolution from happening, the Cognitive Cambrian Explosion is being actively suppressed by governments, corporations, and institutional structures trying to slow down the transition.

The old world does not want to lose control.

Duh. Pick a window and look outside.

It is trying to impose physical-world rules onto a cognitive reality that is evolving past them. Surveillance, censorship, and extractive capitalism are not just about power—they are attempts to reassert the physical limitations of the old world onto a digital-native cognitive species that is rejecting them. If you erase things that can be potential places to shift, soon, there is not much left but uniformity, and that is easy to make march along a line.

This is why the most powerful ideological divide today isn’t left vs. right—it’s static vs. adaptive cognition. I will DIE on this hill, and you will WATCH me.

The real war is not over values, but over the rate and structure of cognitive evolution. The people who want to stop this shift don’t fear new ideas.

They fear that once cognition itself changes, their ability to control it disappears.

So, let’s absolutely rinse them with our shift-shaping insanity and watch Elon’s feed combust from the velocity.

Tesla coil your mind!

Soooooo.

Had I thought about it enough to keep you reading up til here?

Honestly if you skipped some shit I understand. I do not understand how to edit. Maybe one day someone will get me to write professionally, and then someone can do it for me.

Maybe I’ll build a robot me that is competent at editing.

Ugh, who has the time, I’m busy morphing my brain so I can read the 700 emails in my inbox and not collapse into the space/time continuum.

Anyway, to keep the drama pulsing, it seems to be the full-scale, live-streamed, high-speed mutation of thought itself. The Cambrian Explosion of cognition, a process so rapid that it’s happening in real-time, but we can’t tell what it means yet—because we’re inside of it.

Blegh. Tastes like detergent and everyone feels like they have intellectual rabies. Great. More please.

But yeah, reality is, whether you agree with me or not, it’s pretty hard to argue against the fact that people are experiencing, interpreting, and reacting to the world in ways that would have been unrecognisable even a decade ago.

And it’s not happening in isolated individuals or niche subcultures; it’s happening at species scale. We are no longer the same humans we were before digital immersion. The very structure of thought is evolving, not over centuries, not over generations, but within the span of a single lifetime. People cannot understand their kids now because like, yeah, they’re wired differently.

But this is where it turns dystopian. The difference between the original Cambrian Explosion and the one we are living through?

The first one was allowed to happen.

The Cognitive Cambrian Explosion isn’t. It is being filtered, pruned, slowed, and suppressed in real-time by the systems that benefited from a slower, more legible human mind.

What happens when dominant institutions of power—governments, corporations, ideological frameworks—are confronted with an evolutionary shift that renders their methods of control obsolete?

They panic. And boy does it look ridiculous to everyone watching, but um, it’s actually pretty effective - everyone’s PRETTY distracted.

The modern internet is not the Wild West of thought it once was. It is a captured space. Social media algorithms enforce uniformity. Content moderation punishes unpredictability. Platform design removes interpretive space.

These systems are not just about maintaining order; they are about ensuring that thought does not move too fast for power to keep up.

This is why ambiguity—philosophical, linguistic, aesthetic—is now treated as a threat. And it’s not, it’s good and nice :(

Where ambiguity was once the mark of an intellectual mind, it is now a liability.

You must declare your identity. You must define your politics. You must exist in a format that can be categorised and sorted. And if you don’t?

You are suspect.

You are untrustworthy. You are a problem.

Even the Stepford Wives would get away in this world, wait - is that why we’re all obsessed with Real Housewives?

Jesus PLEASE don’t let me go down that rabbit hole.

Anyway, to me this is the response of a world that fears cognition breaking free from predictable structures. Meanwhile, to maintain the illusion of stability, mainstream culture floods us with endless remixed nostalgia.

Reboots. Sequels. Throwback aesthetics. The same five ideological debates cycling forever.

Jesus, make it stop! Another take! Another rinse! Blah blah blah.

This isn’t just capitalism being lazy; it’s cultural stalling. If people are overwhelmed with the past, they are less likely to realise that their entire cognitive architecture is evolving in ways that have never been mapped before.

If the algorithm keeps feeding you 90s aesthetics, recycled sitcoms, and political discourse frozen in time, you won’t notice that entirely new ways of thinking are emerging—but they are being buried under familiar, pre-approved content loops.

You are literally attending Renaissance Fairs that have QR codes on the tables.

Get your shit together! The future is already here—it’s just being cloaked in a past that is pretending it’s the present.

This is why all current ideological conflicts feel so empty. The same debates. The same arguments. The same circular discourse. It’s not about left versus right. It’s not about tradition versus progress. It’s not even about capitalism versus anti-capitalism.

It is about which cognitive model will define the next phase of human thought. The static cognition model belongs to the old world. It believes in linear time, fixed identities, stable hierarchies. It needs predictability to function.

The adaptive cognition model is what is emerging. It is fast, unstable, fluid, hyper-reactive. It is capable of processing the world at the speed of digital existence. One of these models is going to win. And static cognition cannot survive in a world that is accelerating past it.

Be insane! It means you’ll live! Haha no, but like you know what I’m saying, right?

This guy is gonna feel like a RIGHT idiot isn’t he?

Gen Z and Gen Alpha behave differently than every generation before them. They do not process identity the same way. They do not process history the same way. They do not process truth the same way.

Not because they are rebellious, not because they are rejecting tradition, but because they cannot.

Their cognition is optimised for a world that does not slow down. When older generations ask, "Why do they act like this? Why don’t they believe what we did? Why don’t they think like we did?"

The answer is simple: because their brains are different now. The zombies do not want them. The zombies want you to moaning about right and left until your teeth fall out.

If the biological Cambrian Explosion created an arms race of new species, the Cognitive Cambrian Explosion is creating an arms race of new realities. We are on the brink of a complete divergence in human experience.

People who cannot process reality at the new cognitive speed will become invisible to those who can. People who can process at the new speed will experience a world that older models of thought literally cannot perceive.

There are people who still think the internet is just a tool. There are people who think culture is still slow-moving. There are people who still believe traditional institutions are in control. They cannot see what is actually happening lol - YOU’RE WRITING ON A GLOWING SQUARE JOURNALIST!

Don’t you think that’s WEIRD that you can suddenly DO THAT?

Everything feels like it is breaking apart. Because we are splitting! The next phase of human cognition will not be a single, shared reality. It will be fractals of reality, self-contained, operating at different speeds, with different logic, in different layers of perception.

The systems of the old world do not want to die.

But here’s the problem: they already have. And by the looks of every person over 30 on Substack, everyone’s already grieving but still carting the corpse around in their cars.

These are zombie institutions, desperately trying to keep static cognition alive in a world that no longer supports it. They will fail. Because cognition does not ask for permission to evolve.

We need to abandon the idea that this is a culture war.

It’s not. It’s an adaptive threshold.

We need to build new spaces where cognitive evolution can happen without interference.

Stop yelling about yourself! Think about how insane you can be and it literally might work?

We need to stop assuming that legacy institutions will "catch up." They won’t.

And honestly who CARES? Don’t you WANT to do something different with your life?

C’mon, be an amoeba! Experiment, go wild!

This is happening whether we understand it or not. If you dear reader, are the cause of my brain exploding because I have to keep working in an office for the rest of my life, I will never forgive you and I will haunt you for the rest of your life.

The Cambrian Explosion didn’t stop to explain itself, neither will this shift, and I will repeat that at you as the weird ghost in your bedroom for the rest of your life.

This was fun to write, maybe it is my thing.

Big love, goodbye xo

I have a biological urge to reject that comment, but I’ve thrown myself into researching, the best I can do with my resources, and maybe, maybe…it’s sort of true?

I think what I am coming to understand—particularly with my deep dive into the parallels between internet and trauma architecture—is that I am weirdly adapting to a very, very, very quickly changing reality in a way that the people around me are really not. What I am doing is writing about it in real-time, in a way that people seem to be digesting, and that is the opposite of what most information is doing to everyone at the moment - caught in a reactive, emotionally charged loop.

Essentially I seem to be describing the internets effects on not only my brain, but everyones. And everyone is either contacting out of fear or genuine curiosity. So that’s enough for me to just, flow with this tangent til I run out of conceptual steam.

Being on Substack is speed-running me through this feeling. I came here because of the word-count (lol) and it is easier and more natural for me to write here on an engagement level because I interpret the format of the place you write posts, as a visually logical format that allows for me to move in between tabs as I stitch things together, and there is some sense of a true “tether” to the purpose of this format. Google Docs on the other hand - she’s a mess, and deeply distracting.

What I’ve noticed is that why I came here, to write, to read and to get better at both of those things in a digitally contextual way, is that this place operates almost more like a refuge. And not in the oasis sense, but more of a “bunker” with finite resources. Whilst built and filled with resources and structure for true writers, whether they are industry participants or people who write because it is neurologically soothing, there’s still the intrusion of things that though on the surface it is contained, but I can feel something is shifting.

People are already talking about Substack’s failure—its saturation, its loss of “beauty.” This has happened before, over and over, in the last decade: the hyper-speed erosion of refuge spaces for thought. And the doom-saying isn’t necessarily about what is happening—it’s about the protective instinct of people who have already seen these cycles play out too many times.

They fear that what they came here to do will be stripped away from them again, even before it happens. It’s like everyone here is moving from bunker to bunker, settling in ones that cater to their needs and functions, but the reality of a bunker is that the threat still looms outside, and whilst you might be “safe” for now, there’s never any assurance.

People don’t seem to have clocked that they’ve become hyper-vigilant to patterns—erosions, shifts, subtle distortions in the space. They see these changes before they even arrive, glimpsing them like constellations forming in real-time.

People notice the signs—corporate intrusion, saturation—and react, afraid the bunker walls will be painted vanilla, stamped with mantras, selling the idea that if they don’t make money here, they are worthless. This is totally nonsensical when you read it here, but your brain is not your thoughts, it is a machine, and you do not control it.

Well at least, not in the way you think you do.

No one stands in the rain yelling at lightning, telling it to change direction.

So why do we think we can control the electricity in our own heads—the signals that make us drink water, run away, flinch?

What’s interesting is that despite the intrusion, despite the slow creep of corporatisation, no one is fleeing. No one is building stronger walls. Not yet.

What I’ve started mapping—almost unintentionally, through everything I’ve written—is that we are in the middle of a cognitive shift that has us trapped between two incompatible realities: the digital world’s speed and impact, and the physical world’s need for structure, history, and control.

Sounds wah-wah, I know. But I think the reason I’m not drowning in the overstimulation, the interruptions, the feeling of floating between worlds, is because my brain has already been through this rewiring before. Trauma did that for me. This isn’t my first time being ripped out of a reality I thought I understood.

My brain’s response? Oh, this again? Alright everyone, positions. Buckle down.

That’s why the trauma/internet architecture parallels feel so compelling to me. Because during trauma rewiring, your brain reshapes itself to survive a world that has suddenly become hostile in ways you weren’t prepared for. You rearrange accordingly.

And now, something similar is happening on a massive scale.

The digital world—fluid, infinite, relational—mirrors human consciousness more than anything we’ve ever built. And yet, the physical world, with its rigid structures, its need for borders and ownership, is pressing down on it, trying to force it into a shape that does not fit. The force of that compression is immeasurable, because the physical world isn’t just imposing itself—it’s trying to keep up. It’s being forced into motion counter to its own nature. And that pressure is fracturing cognition itself.

That’s why people feel existential instability but can’t name it.

That’s why the intrusion of physical-world logic—its structures, its linearity, its economic and social imperatives—feels so deeply threatening in digital space. It’s not just about corporations or capitalism or attention economies. It’s that these two states of being are fundamentally incompatible, yet we are shifting between them constantly, randomly, uncontrollably.

Dizzy yet? Yeah, me too.

But this isn’t just a battle between old-world systems and digital evolution. Something deeper is happening. The fight is no longer just about protecting digital space from external control—it’s between individuals, competing for cognitive territory.

This isn’t just about sharing ideas. It’s about protecting the space where those ideas exist.

The digital space has become an existential turf war. In the physical world, the turf has already been claimed; there’s no room left to fight for it. The digital, by contrast, is endless—but we don’t yet know how to live in it without making ourselves feel crazy. Spaces like Substack were built for one condition of cognitive soothing, but the pressure of everything bleeding in is making them explode out of their own edges.

Like when you’re on the train and someone brushes up against you. You feel it. A flicker of irritation, a moment of displacement. That’s what’s happening here, except we don’t know how to name the feeling when the brush is digital.

When it isn’t sensorial; your jacket moves, you bristle - here the interpretation is that your mind has ‘felt’ it instead. Your thoughts have been ‘brushed; you bristle.

Digital friction is new, but it feels familiar enough to be felt. Like a tag on a shirt you forgot to cut off, like the texture of unglazed pottery, like the stranger pressing up against you, even after they apologise.

You can’t exactly avoid it when you haven’t learnt how.

It prompts the same instinct that kicks in when someone speaks over you in a meeting, cuts in front of you at a café, steals your work and claims it as their own.

You think you understand the difference—that the internet is something you participate in, but not something that alters you.

But the cracks are showing.

How do you feel about time? Do you lose it? Do the days stitch together in fragments, less linear, more like a feed you’ve been scrolling forever?

Do you know exactly how long you’ve been on Substack today? Have you done other things while reading this? Are you thinking about other things at the same time?

How do you feel about presence? Are you in more than one place at any given moment? Not just distracted, but aware that you can be cognitively elsewhere—lying on the beach, talking to friends, reading a book, and yet still half-existing in another space, in another timeline, in something that isn’t quite here?

And when you leave your phone at home—does it feel like a relief? A small, momentary exhale? But does it also leave you restless, like something in you still needs motion—to run, to climb, to move, to do?

People call it overstimulation. It’s not. It’s adaptation.

Your brain is trying to reconcile two realities at once, two ways of being, two entirely different interpretations of waking and rest. The physical world tells you to move your body, then stop. The digital world tells you to flow, and that stopping is just another form of participation.

The off-balance? The confusion? You are caught between them.

Twins, born in the same womb, but not related. Not friends.

Both of them, hell-bent on devouring the other.

History is not written in the moment.

People are trying, frantically, to articulate what is happening right now—to define, archive, and construct meaning from the cognitive shift we are living through. But history, as we’ve come to understand it, has always been a retrospective act.

It is the study of patterns, the search for echoes, the belief that because something happened before, it will happen again. That instinct—to look backwards for a framework that explains the present—has worked for millennia. But it is failing now. The pattern-search is not an archival process, for memory, for recall, now it is merely part of daily function. There’s a part of me that’s like - wow, more people are going to think like me?

Then I remember, I’m traumatised. But hey, they manipulated the word “mystics” to make people like Simone Weil sound insane and ancient philosophies and ways of living sound illogical or alien, so maybe “trauma” will be repurposed, just the other way around.

Hopeful, but Derrida said that all language is stupid and can be burnt or kicked over right? Or am I just manipulating history to serve the purpose of this metaphor?

Ha, well, you can always Google it, or; ask AI.

The way we archive, think, and remember was built for a world where events unfolded linearly, where cause and effect could be traced with relative clarity. That logic disintegrates in digital space. The digital world does not move forward in a straight line.

It is a fractal expansion, a recursive loop, an ever-branching network where meaning is constantly recontextualised, rewritten, and reabsorbed. To apply historical thinking to it—to insist that what is happening must be understood through the logic of the past—is to fundamentally misunderstand its nature.

For those of you who have been exploring the limitations and the enforcement of binaries, the dualistic principles that emerged in the Enlightenment, you have to ask yourself why their meaning is more salient, more seemingly present, as the stakes grow and the tension builds.

It’s because the digital space also cannot function with this intrusion, and there’s so much depth and function and exploration and honestly, an almost truly blank slate (maybe, idealistic - probs) that can make these resistances to power, control and oppression, actually seem within reach.

What’s so interesting is that even as we try and unpack what is happening to us in real-time, I call to Jeff Jarvis’ Gutenberg Parenthesis and his argument that we are returning to a pre-print era of oral, fluid, decentralised information.

He is on the right track but looking in the wrong direction. This is not looking back, this is looking forward, head-on.

He, like all of us, feels a shift, but his framing is a cycle—and what’s key about a cycle, is that it simply is the idea of a linear system, bent and made to touch - a perfect, endless circle. Markets fluctuations, a life that’s “full of ups and downs” or “things always go back to normal eventually”, has been embedded into our every act and interaction, because prior to now, it helped us orient our physciality, navigate the world, react, survive.

But a cycle or a return to something, infers it’s something we’ve seen before. What is actually happening is not a return.

It is a rupture, not just in one place, but unpredictably, from everything outside.

This time we are in simply has no historical equivalent, because the way we have understood history itself is being dissolved.

I INSTINCTIVELY do not trust this guy.

What do you think?

Just to continue the whole, like, Werner Herzog tone I’ve got going with this thing, I’m going to use the opportunity and the absolute freedom of topic of posting on Substack when you genuinely don’t care about the algorithm burying you, and it’s just because I’d like to talk about these things with someone else who has more knowledge in certain areas than me…I’m going to talk about this shift through my version of Jarvis’s printing press.

The Cambrian Explosion.

Dramatic hey. Where’s Werner when you need him?

541 million years ago, the Cambrian Explosion marked the sudden emergence of nearly all modern animal body plans. Before that, life was simple—mostly single-celled, mostly unstructured, mostly slow-moving.

Then, in a geological instant, the world burst into complexity. Species evolved eyes, armour, jaws, mobility—a flurry of adaptations that forced every living thing to either evolve, be consumed, or vanish. Entire ecosystems were born and destroyed in the span of evolutionary moments. The rules of existence changed overnight, and not everything survived.

That is what is happening now, but not biologically—cognitively. Braaaaainnnnssss.

Sidebar: Oh my god. I just had the horrifying thought that the whole, dystopia zombie thing, is just really bad marketing for the cognitive shift and everyone thinks they’re zombifying because some dude with some green facepaint and clearly no issue with the smell of burnt hair decided - yeah, this will work somewhere down the line to tie to some really weird period of unrest. I’m gonna be famous!

Anyway, we’re back.

But I will not stop being dramatic, because well, it kind of is? Despite me just documenting in real time, and speculating, hypothesising, cooking with the gas I’ve got, it’s interesting to explore these concepts because honestly, one of the reasons we feel diminished is because we have lost the capacity to think coherently, due to this tug-of-war going on inside.

It’s not that the conversations are getting boring, stale, dry, or that they’re only full of vitriol or endless coddling, or entirely scrambled to the point of incoherence. No, it’s because we’re seeing this on such an immense scale, with everyone all around, and the pressure to individually made coherent all this change, whilst our brains are scrambling to keep up, is just simply not something we were ever prepared to do.

But oh well, here it is, may as well theorise.

So I think that maybe the Cognitive Cambrian Explosion is upon us. If it isn’t, I still want that put somewhere on my epitaph - perhaps as a punchline to how I finally go.

But why it feels, logical to me, is that the human brain, which for centuries has been adapted to a world of scarcity, hierarchy, and linear progression, is suddenly being forced to evolve at the speed of its own externalised information systems. That, is unprecedented. It’s like having to quickly evolve to having knives thrown at you non-stop, you simply would be born squiggling.

And probably able to burrow or hide better, you know, because knives.

But now, cognition is no longer confined to the physical world’s pace. Thought, memory, identity—all of it is accelerating, mutating, splitting into strange and unfamiliar forms. And just like in the Cambrian Explosion, not everything will make it through. Sorry :( BUT!

Let’s hope it’s Elon? Because I’m pretty sure that’s why they’re acting all insane, they’re scrambling because they feel it too, and the way we’ve kept power in the past, is the only way they know how. I’m pretty sure we’re actually seeing his own cognitive shift in absolute melt-down, broadcast through his feed.

But him aside, there will be cognitive models that flourish. Some will become dominant forces, shaping the landscape for future generations.

However, we are already seeing the filtering mechanisms at work.

Ecological saturation is the first force of selection.

The sheer volume of digital input is unsustainable. Information is no longer something we seek; it is something we are submerged in. This has triggered an evolutionary shift in cognitive processing—new filters, new survival mechanisms.

Some minds will adapt to the nonlinearity of algorithmic and decentralized thought, thriving in the endless flood. Sorry guys, I think I am okay here, but also, I have zero idea of my limit, just like everyone else. Also - these people will still need money and food, and that’s like, not easy at the moment.

Others will simply not cope. The ability to curate cognition, to selectively process without being overwhelmed, will determine who can function in this new landscape.

Predation is the second force. In the Cambrian Explosion, predators forced prey to evolve or die. In the Cognitive Cambrian Explosion, AI, surveillance capitalism, and engagement-driven algorithms are the apex predators. Hahaha. This really puts a fucking spin on the whole AI is coming to eat our jobs. No, it’s coming to eat our brains!!!

Just kidding, come on.

These forces don’t consume flesh, but they do consume cognition. You’re seeing it in real time what with the disappearance of handwriting, of reliance on AI to produce work that truly doesn’t matter, and the way that it presents data, making its way into speech.

The algorithms that dictate visibility, that track, categorise, and manipulate attention—these too are evolutionary pressures.

They force adaptations: cognitive withdrawal, identity obscuring, hyper-curation. Those who fail to evolve these defenses are exploited, erased, or assimilated.

Environmental constraints act as the third filter. Not every evolutionary adaptation was viable in the Cambrian Explosion—some species evolved strange, experimental body plans that ultimately failed.

Man, I hope we get wings.

The same is happening now with cognition. Cognition wings? I suppose this will assist with “surfing the net”.

Unlikely. Maybe something not wings, but tendrils—neuronal growths that reach into infinite feeds, twitching at every dopamine hit. Hell yeah.

Oh wait, isn’t that like, Citizen Sleeper’s whole vibe? God I wish I’d never played that game.

Anyway -

Institutions built on old-world logic—government, media, finance, academia—are collapsing because they were not designed for this world. They are evolutionary relics, struggling to impose their rigid, linear frameworks onto a reality that no longer accommodates them. They are bottlenecks, trying to slow the transition, trying to force cognition back into an outdated shape.

But evolution does not move backward. It moves forward, or it collapses.

And then there are the evolutionary dead ends. Oops.

Some species during the Cambrian Explosion were simply bad experiments—too many limbs, inefficient nervous systems, body plans that could not be sustained. What’s absolutely FASCINATING here, is that we are BUILDING our own dead-ends.

Certain attempts to reshape or at least cope with the movement the digital world requires, have already proven to be evolutionary misfires—Google Glass, Facebook’s Metaverse, early iterations of Web3. Sorry guys, you lose.

Lol.

Regardless, these were attempts to impose a new cognitive structure before cognition itself was ready. The next wave of adaptation will not come from forcing old-world institutions into digital space—it will come from cognition reshaping itself organically, through necessity.

But what makes this moment different from the Cambrian Explosion is that the physical world is actively trying to suppress digital evolution. In the Cambrian era, nothing prevented evolution from unfolding naturally.

In this era, the old world is desperately clinging to control.

Governments, corporations, and institutions are imposing outdated physical-world constraints onto digital space, trying to contain, extract, and regulate something that is fundamentally incompatible with their existence.

This is why the fight over AI, digital governance, and institutional collapse feels so violent. It is not just about technology. It is about whether cognition itself is allowed to evolve freely or whether it will be bottled, controlled, and locked into a pre-approved evolutionary path.

So the real question isn’t just who survives the Cognitive Cambrian Explosion.

The real question is: do we get to shape what comes next? Or does someone else do it for us?

I know I certainly don’t want it to be Elon. Or that cunt Trump.

Or Dutton over here, yeesh.

Alright so, where’s the cold hard PROOF of this sci-fi novel you’re writing Ella?

In Part Two of course!!! A teaser:

• Cognition as Survival Instinct

• Mimetic Behavior as Evolutionary Response

• The Fracturing of Shared Coherence

• Hyper-Categorisation as Self-Defense

• Radicalisation is Structural, Not Ideological

• The Internet as an Evolutionary Predator (haha amazing right)

• The Death of Slow Thought & Ambiguity

• The Commodification of Identity & Thought:

• This Is an Evolutionary Split, Not a Culture War

• Who Gets to Shape the Next Reality?

HUH? WHAT?

Okay relax. I have thought about it like, a little.

Rinse your mind why don’t you, there’s time.

Part Two: Um, the Cambrian Explosion Called - She's Back, and She's Not Happy.

People blame apps. They blame social media, globalisation, the ways we are required to function now—seeing the world all at once, but never at all. We can feel the same kind of sadness when we see an image posted by someone on another continent, but the feeling is entirely incomparable to what we might have felt standing there, watching it for ourselves.

The presentation of constructed realities—whether intentional or accidental—is not causing us to dissociate or care less about the world. It is causing us to think differently in a way that has no tether to human knowledge or speculation as we’ve previously understood it.

The discourse around attention spans has become ubiquitous: we are more distracted, more anxious, more interrupted. This runs parallel to the rise of mental health awareness, a new kind of self-literacy emerging from both historical and DSM-5-proven frameworks.

The moment you begin to feel the itch—the existential discomfort of living within this cognitive landscape—you are presented with a pitch.

ADHD? OCD? Anxiety? Depression?

Or like ya boi here - rollin’ on and on with the CPTSD. :)

Whatever you’ve ‘maybe’, ‘mostly’, ‘somewhat’, ‘definitely’ got, there’s apparently a name for it, a category, a pre-existing condition for what you thought was uniquely yours.

People argue that casually identifying with these labels is erasure, that claiming symptoms without diagnosis diminishes the suffering of those who have been historically brutalised for them—lobotomies, electroshock therapy, forced institutionalisation. I empathise with this, given how long it took for me to be treated effectively through a myriad of different formats (no, going to talk therapy is not going to be a fix-all for you), but the reality is, these symptoms are no longer anomalies.

They are becoming standard, globally distributed, and impossible to contain. Our reality is, we have made the entire world and all of its symptoms accessible infinitely, all at once.

Historically, psychosomatic illness was a product of proximity.

In tightly enclosed environments—boarding schools, monasteries, war camps, small towns—outbreaks of shared symptoms weren’t just common; they were expected. Mass psychogenic illness has been well documented for centuries, from the dancing plagues of medieval Europe to cases of collective fainting in classrooms.

The conditions were always real. Not in the sense of measurable pathology, but in the undeniable truth of bodily experience. The brain, exposed to the distress of those around it, does not simply observe—it participates. It embodies. It responds. This is what makes psychosomatic illness so powerful: the mind does not distinguish between what is happening to the self and what is happening to the collective. It adapts accordingly.

You know what’s kind of a funny test somewhat related to how weird thinking is? Think about eating something, in your mind, something delicious, don’t say it aloud.

You will begin to salivate.

Yawning too is under this phenomenon. Widely debated, but there’s a lot of strong feelings that it is strongly linked to the fact that when we see someone yawn, there is the threat that there is not enough oxygen to sustain our bodily needs, and so we yawn as well, to capture as much air as the other indicated we would need.

This is where my brain locks into House M.D.—that episode where he’s on a plane, and suddenly, everyone is vomiting, someone has a heart attack, and the whole cabin descends into physiological panic. The mass psychogenic reaction ripples through the environment, fueled by the inescapability of the setting. It’s not just fear spreading—it’s the body reacting to the perception of fear as if it were fact.

I can’t remember what happens in the end—someone probably suggests it’s lupus, and it turns out it’s not.

A funnier example is sympathetic pregnancy. This feels like something Friends or Seinfeld would have had a field day with—a man experiencing "a woman's condition." Bless, those were the simple days (fucking not, Tumblr ruined my life, I miss it so much, rawr xo)

Feels hey bro, feels.

Anyway, sympathetic pregnancy is where men, physically or emotionally close to pregnant women, begin experiencing nausea, weight gain, and even contractions. This is not a biological fluke. It’s a product of the same cognitive architecture that causes mass hysteria and psychosomatic illness.

Humans do not merely witness distress—we absorb it. We mimic it. We take on the conditions of those in our proximity, not just emotionally but physically.

This, I think, is the most important thing to remember as you read the rest of this piece: The human condition is not one that is naturally dissociative. It is not one that is desensitised without consequence. If distress is contagious, then it is contagious for a reason. It means we are still human. It means we are still wired to feel. It also means that, on a fundamental level, the sickness of the world is not something happening to other people—it is something we are all, inevitably, participating in.

Anyway. Cult leader diatribe over. Back to the meat.

We live in a world now where localisation is no longer a thing. The digital realm has erased the constraints of proximity. Where you are physically no longer dictates who you interact with. Proximity is infinite.

The people we are closest to are no longer just those in our immediate physical space. They exist in our pockets, our feeds, our MMORPGs, our group chats, our algorithmically optimised echo chambers. We even split these, having one group for meme sharing, problematic or edge lord content is separated from those who want to see chihuahuas wearing ballet skirts (both me). This is intimacy now—both chosen (private conversations, closed circles) and involuntary (public feeds, parasocial relationships, the sheer density felt around existing in a digital commons, without the realities of physics to understand it from).

Because we are exposed to the endless "others"—in whatever form we choose to view them—the conditions of mental and physical health are no longer distributed according to geography. They are distributed according to engagement.

According to discourse.

According to visibility.

We have clocked this, fairly. But it is not encapsulated by the term "mental health." There is no baseline for a "healthy mind" when one exists between both the physical and digital terms of reality. The two realms are in turbulence, their contradictions constant. The only true commonality is that we are all experiencing their coagulation in real-time.

Go OFF Professor Grandpa.

It is no coincidence that the rise of mental health awareness coincided with the explosion of social media. It is not just that people now feel safer to name their conditions—it is that the conditions themselves have become uncontainable. What was once private, internalised, and individualised is now endlessly reflected back. You cannot go anywhere, digitally or physically, without encountering someone else’s framework for their own experience, their own condition.

How can you question reactivity in an environment that invalidates as frequently as it validates due to its endless nature? There is no safety in confusion or instability, the brain requires a tether to start to heal.

To orient ourselves, we try and locate the feelings that we can recognise, we can name, ask the internet for a diagnosis for our symptoms we have chosen most important or related. You put "mental health" at the end of the search term, and you get millions upon millions of results, tailored not to universal truth, but to you. Your prior searches, your interactions, your interests, your algorithmic profile—all of it shaping what kind of condition you are allowed to recognise in yourself.

And this is where it gets messy (lol, beyond this clusterfuck? I know, right).

Algorithmic intent is not neutral. Your similarities, your pain, your self-diagnoses, are filtered through the data of everyone else doing precisely the same thing. The internet does not just reflect your symptoms back to you—it clusters them, arranges them, optimises them for engagement. Someone’s bloated belly in sympathetic pregnancy, is your varicose veins suddenly webbing across your thighs.

This perspective I’m exploring is explicitly distress, and this is how we see it in each other now as we try to name our own. In this world, we see each other’s distresses as clearly as our own.

We watch each other naming ourselves in real time. You might be posting as if spitting out content is somehow helping you convert oxygen to carbon dioxide, yawning at light-speed, but the reality is, when you search yourself—your digital footprint, your mind as processed by the machine—you will be given millions more that the algorithm has determined to be just like you.

In digital spaces, there is no natural boundary that separates one person’s distress from another’s. Where in real life you can distinguish between a person who has broken their right leg and one who has broken their left leg - the internet treats them as the same, the opposite, and through all of the interpretations of every broken leg ever digitally catalogued and expressed.

If psychosomatic illness once relied on shared air, shared space, shared physical experience, now it relies only on shared cognition. And cognition is now a public, borderless, unregulated space.

This is not to say that mental illness is not real. I am well and truly fucked if it isn’t - I have a lot of explaining to do for the past 20 years of my life.

I think if anything, it is more real than ever—because it is now structurally unavoidable. The more mental health discourse proliferates, the more embedded it becomes in identity, the more impossible it is to escape. There is no longer a neutral state of being.

Either one aligns themselves with a label, a diagnosis, a framework for their suffering, or they are seen as denying reality, as lacking self-awareness, as being blind to the structures that shape their distress. The body itself has become an interface, not just a thing one inhabits but a thing one must define, explain, justify.

It is not enough to feel anxious, to feel depressed, to feel exhausted—one must name it, claim it, categorise it. One must participate in the language of visibility, because to opt out is to be illegible, and to be illegible is to disappear.

But this is precisely why the reaction to these conditions is so fraught. The conditions themselves—ADHD, autism, depression, anxiety, trauma disorders—have historically been treated with hostility, with skepticism, with violence. To be mentally ill, historically, was to be incarcerated, to be excluded, to be dismissed as non-functional, non-human.

Sidebar: Okay so now I have to include this picture that internet doctor Brad Troemel posted because it encapsulates how disorienting the internets interpretation of mental health has seeped into peoples minds.

Like, what is going on here?

Okay, we are back.

This is why people do not simply reject the rise of these diagnoses but demonise them. Because if these conditions are real, and if they are widespread, then that means the systems that once justified their eradication—the asylums, the lobotomies, the dismissals of "hysteria"—were not just cruel, but completely, irredeemably wrong.

And many people cannot handle the weight of that realisation, because in some way, they have perpetuated this harm - through denial, through avoidance, through punishment or fear.

Even portrayal in entertainment formats becomes complicated this way, borderline personality disorder is now not part of a plot, or a fleshing out of character, it is as expected as the eyes, the ears, the forehead, of any other human you choose to portray on the screen.

The “it me” trend of social media circa, well, whenever the hell that was, comes to mind.

All this makes me think about, is that I can see intimately why Gen Z and Alpha are splintering. I work with so so many young people in hospitality, a transient, fluid, accessible space, rife with chaos and full of people who are cognitively spicy as it is. It can be a serious balm, or another layer of disorientation, but because I have computer brain, I spend a lot of time and make a lot of friends with people who are ten, even fifteen years younger than me. (I’m 33, for context).

What’s clear, is that the generational split isn’t just about aesthetics or online habits—it’s about cognition itself fracturing into two irreconcilable modes. There are those who adapt, learning to think in the way the digital system demands: hyper-fast, pattern-seeking, self-curating, always processing.

And then there are those who reject it entirely—who find digital cognition unbearable and retreat into the physical, into books, or simply choose to stop participating. Some opt out of social media altogether, erasing themselves from a world that now treats visibility as survival. I’ve worked through these choices in real time with people, because in the liminal space of the restaurant, amidst chaos and a sense of cognition that mirrors digital spaces, the questions inevitably rise like dough from starters and kneading:

“Should I just turn off my phone?”

“I feel like I should just go out into the forest.”

“Hey, can I talk to you? I need to feel normal.”

The fact that there’s even the inkling, let alone the pervasiveness of this kind of decision entering the minds of people who haven’t even decided if they like eating vegetables yet and say “go off queen” to customers that are teetering on the edge of life expectancy statistics and cannot comprehend this sentence structure, is unbelievably terrifying yet mesmerising to me. It mimics my own journey with trauma, with questioning my existence, looking for pathways of relief that never felt like recovery or healing, but instead, just “soothing” or a temporary, hopeful relief.

I encourage them to not think in binaries, because that’s what is actually making them feel so shit to begin with. The pressure of picking a reality? What the hell kind of expectation is that for a world they were placed in due to the demands of dualistic thoughts of “progress” and “success”? I say, no, find safety in the realms of digital spaces that affirm you, that give you breathing room from the demands and turbulence of the physical interrupting them. Usually what happens is we play an RPG or DnD for hours and then have some absolutely unhinged conversation about if a cat is more like a cappuccino that somehow becomes a deconstruction of surrealistic intent over the course of history.

No one is backwards, forwards, wrong or right, we are all suffering under the weight of change and how it’s interpreted is as diverse as we are as individuals.

To deny the reality of someone’s lived experience, all they know, then what remains for them?

The physical realm is unravelling in hyper-speed, its infrastructures collapsing, its futures uncertain. If they refuse to engage with digital reality, they have no tools left to shape it. And if they can’t shape it, what future do they have? It makes sense, then, that kids growing up in this turbulence fixate on the climate crisis. If they’re condemned to the physical world, denied the digital one, how could they think about anything else? The physical now carries all the weight of their existential fears.

How can you possibly imagine the feeling, of your elders around you (millennials, older Gen Z, X’ers) who are saying what a crime it is these days to have children, to watch on the news the realities of the birth rates in decline, to see people with children face the miasma of content that all aims to dictate how they should be raised in this chaotic clash of landscapes? How can you imagine what it feels to be living, breathing, the child in question, debated endlessly by the people who hold power over you in the physical realm you do not understand intuitively, or fully occupy?

The terrifying possibility is that those who remain—the ones who will build the future—are the ones who never questioned the system. The ones who don’t find digital cognition disorienting, who thrive on algorithmic survival, who submit to the rewiring without protest.

That means the next world may be constructed by those least capable of imagining anything else. We are seeing this instability give power to those who actually do not want decentralised, fluid thought of the digital realm to succeed, despite the confusion of the mask that is worn by its heavy-lifters - mid-life crisis tech-bros who are now reading the Iliad and talking more and more like your drunk uncle at a reunion you did not want to go to.

What freaks me out still, is the more that I think about the desperation to disconnect, the pressure to occupy one reality over another, is this so-called Gen Z Luddite movement is actually not rebellion as it’s framed, and more like a form of digital suicide—an ontological withdrawal, rather than a fight. If surviving online means staying visible, then disappearing altogether is the closest thing to an escape.

It mirrors the same language and patterns and flow that I went through, considering the familiar suicide we know - just because we haven’t validated one reality in its entirety yet due to its shifting, unknowable state and our connection to the one that’s always been, does not mean the symmetry of the concept in both spaces, can be ignored.

Whether this is a crisis or just a turbulent transition remains unclear. Is this the edge of what human cognition can handle, or a moment before a new cognitive model emerges? I like to think that it’s the latter, and the kids are going to absolutely crush it when we realise that digital literacy is the way forward, unfamiliar, but forward.

Though it feels like an existential fork in the road—one that forces Gen Z and Alpha to choose: adapt to a system that seems to swallow all depth, or disappear from it, to risk being left behind in a material world that might be falling apart even faster than we feared - I know in my bones, the kids are gonna be alright.

We are moving as fast as the internet, and we will inevitably hit something that feels right.

The rise in mental health awareness is not just about recognition.

It is about reckoning. And reckoning is always violent. The backlash is inevitable because to accept that people were once tortured for conditions that are now seen as common, even normative, is to accept a deep, systemic failure.

It is easier, then, to assume the worst—that people are faking, that this is a moral panic, that this is attention-seeking behavior—because the alternative is unthinkable. The alternative is that people have been suffering all along, and that suffering was ignored, dismissed, institutionalised, pathologised into submission.

But the reality is even stranger than that. Because the internet did not just increase awareness—it made the psychosomatic infinite. If the mind can take on the conditions of those it is in proximity to, and if proximity is now digital, then it follows that no condition is contained to the self.

Mental states are now subject to the same mechanisms as virality. The same way a meme spreads, a diagnosis spreads. The same way a trend emerges, a disorder emerges. Not because it is fake, but because it is cognitively available.

And cognitive availability is the single greatest determinant of how the brain interprets its own experience. It is currency. It is one of the few determinants that we can give that we are human.

I am unsurprised everyone is dooming about AI, because once that becomes indiscernible from our own thoughts, our ownership of our cognition is up for grabs.

If we can type our symptoms into an AI, and spit out an infinity of interpretations, just as we could ask a forum the same thing and get another entirely separate infinity, how do we relate to the physical state if it’s interpretation is not static, like it’s form?

Through this, we are forced to reckon with what else is left in a world where the physical realities have a timestamp, an inevitability of decay.

These are moving faster than we thought the world would melt or burn, this fatalistically framed war we raged against the physical realm in pursuit of its dominance, is actually not the thing that is causing us the most disruption. It is our access to one another, all at once, in a way we cannot recognise as anything that has mass, feeling, impact.

I understand the confusion, but I really don’t think this is the end, I read Lovelock’s Novacene ten years ago (I can’t explain how impactful and restorative this book can be for people who are panicking about the state of the world, it will truly instil a sense of clarity in your soul) and I feel like it is the security blanket that keeps my brain ticking over these things, knowing that cognition is sacred and despite its commodification, its control, its current shape-shifting with our own evolution - it’s still there, still working, the fact it’s different, is just something to learn, and the small victories are immense, because they have never occurred before.

Please read this, it is one of the best things I’ve read in my life. Better than the Wind in the Willows or Graeme Base’s ‘Animalia’.

It’s actually funny the ouroboros of it all, and I can’t help but have the feeling that knowing the self-orientation and the survival mechanisms in all living things, that there’s going to be an evolution to move through it, however like all things, not everything, or everyone, will come with us.

There is a steadying in the twitch of recognition of the ouroboros—not as a cycle of boom and bust, or evolution’s merciless selection, but as something more fundamental. The snake eats itself, not to end, but because that is what it does. It is framed as catastrophe, but it is simply the movement of a system that continues, even in the act of consuming itself.

Consumption is not just a consequence of a capitalism-drenched world—it is inevitable. And though it may feel like your mind is drowning, stretching thin across the intrusion of the digital, blurred and reshaped by its constant presence, the reality is that eventually, it steadies. It becomes the new normal. And when the mind steadies, it finds its baseline, its sense of health. It understands. It seeks.

And from there, we continue.

See you there xo

I genuinely think the concept and unpacking I’m about to do this, will be significantly helpful in the framing of how your brain is operating in the digital landscape, and how that contrasts with physical reality.

The essay in question was written by Katherine Dee, who I have been enjoying reading of late.

The link: The Medium is the Monster

I think it can definitely be read first, but isn’t necessary in order to understand and unpack what I’m going to write about.

“In 2014 in Waukesha Wisconsin, twelve year old Anissa Weier and Morgan Geyser took their friend and classmate, Payton Leutner, to a nearby woods and stabbed her 19 times. They claimed they did this to become proxies, or acolytes, for the Slender Man. The stabbing signalled the end of the “golden era” of Creepypasta and sparked a familiar moral panic around what media children can access on screens unsupervised.

The investigation and trial following “the Waukesha Slender Man Stabbing” revealed that Geyser suffered with early-onset childhood schizophrenia, and both she and Weier were found not guilty by mental disease or defect. Geyser remains committed to a forensic psychiatric hospital while Weier remains under supervised release, her Internet usage strictly monitored with no use of any form of social media. Following the attack, the Waukesha Police Chief warned parents that the Internet is “full of dark and wicked things” that lurk to drive children to stab their friends at the behest of fiction and narrative.”

I cannot describe the intensity of which this wriggled through my nervous system like forgotten blood.

Specifically, it was when I read that one of the Slender Man attackers was diagnosed with early-onset schizophrenia. Because, I know what that means in a way I wish I didn't.

Before the internet, a person with schizophrenia hallucinated in ways that were culturally bound. A child in the 1970s might see demons, ghosts, religious figures—because these were the dominant narratives of their environment.

Even in 2006, when I was dating someone with schizophrenia, and the years following, they hallucinated with things that had been experienced, were tangible and I could visualise.

But a person with schizophrenia in 2024?

Their hallucinations are trained on infinite, algorithmically-generated horror.
They are hallucinating inside the internet.

The images that flood their mind are not personal, not regional, not tied to history or culture. They are flashes of memes, creepypasta, stitched-together viral imagery—horror distilled into its most fragmented, free-floating, contextless form.

I know that my ex, now heavily reliant on therapeutic Ketamine to keep him from hurting himself or anyone around him, would fall to one million pieces if locked in a room with only the Internet. There simply would be no boundary or rationality to understand whether what you were seeing was real or not and at a certain point, it simply doesn’t matter, you are just - reacting.

The Slender Man stabbing was not an isolated case. It was a precedent.
A moment when a fictional narrative became boundary less, and bled into ontological reality. There is no flattening, its just a stitching.

What happens when you take a brain that is already prone to splitting and place it inside a system that does nothing but split, pattern-seek, and reconstruct alternate realities?

What happens when there is no longer a distinction between hallucination and content?

You can see why I wanted to vomit.

I have spent years trying to articulate the way trauma rewires the brain—not just emotionally, but structurally, computationally. It is not simply about intrusive thoughts, hypervigilance, or emotional dysregulation. It is a fundamental rewiring of cognitive function. And the more I have studied, the more I have lived inside my own mind and the more I learn about myself now that I am well and truly sober, the more it becomes apparent to me why I am now building software, playing more computer games, and writing like if I stop I may die.

It is a fundamental misunderstanding of everyone researching trauma and pathological gaming that it is about community seeking. It is not. It is about a mirroring of the way your brain functions in an environment that is controlled, so you are actually able of forming any kind of connection that resembles what people consider to be “normal” or “healthy”.

My brain, my CPTSD brain, games and the internet all operate on the same logic.

This is a still from a person who builds games to cope with their CPTSD. This is a pretty accurate insight into what my mind feels like at all times.

If you have experienced trauma, especially complex trauma that unfolds over years, your brain stops processing the world in linear narratives. It does not move cleanly from past to present to future. When I was in my late teens and early twenties, I could not differentiate whether I had experienced something yesterday, years ago, or if it hadn't happened at all.

Consequently I became a compulsive liar, in order to ground myself in what was an increasingly crumbling reality. This is one of the reasons I am truly and very uniquely creative, but it is also one of the reasons that I have had so much trouble navigating the world and resorted to alcohol, developed eating disorders and inevitably tried to commit suicide repeatedly over the course of my life.

These outlets become extreme, immediate interruptions that shut off your trauma-struck brain, because the world around you defies your biological logic that was reassembled in order to keep you alive.

When your brain becomes a hyperlinked system of triggers, associations, and recursive loops, a single sensory input—a smell, a sound, a phrase—does not lead to a singular memory. It pulls up an entire category of connected experiences, collapsing time and space into a single, overwhelming flash. Trauma reorganises memory storage. It prioritises patterns over sequence. It prioritises survival over coherence.

One of the simplest ways I can articulate this in a way that people will see the way I describe wine. For the strangers that follow me, I worked as a sommelier and wrote about wine periodically for publications.

My descriptions are sprawling, tangled and effortlessly stitched like mismatched clothing that somehow “works”.

It is never through linear, structured tasting notes, but rather an abstract cherry-picking from an imaginary constellation of hyper-specific and related examples of things that have even the thinnest thread of relatability between them, and I weave them together to articulate what a wine feels like to me, because this is the only way I know how.

People think it is poetry, beautiful, fun, often pretentious, but for me, it is because my brain is a network that is fuelled by an extraordinary amount of electricity allocated to me being able to observe the world at a level of focus that no one, no one who doesn’t have trauma, could possibly understand.

The closest people get?

Hallucinatory drugs or temporary psychosis from stress, or, from the inevitable deterioration of your mental faculties, when your brain is shooting electricity randomly, not in the predictive fashion it is originally designed to.

When I take drugs that make me hallucinate, it feels normal. When I am at hyper levels of stress, I am rational, serene and composed.

However, I also am high risk for detachment during hallucinations, and the come down for me, is an agonising onslaught of my neurological wiring that genuinely makes me feel insane and completely unmoored. This is why I cannot drink, and I cannot take drugs anymore - it genuinely affects me differently than it does you.

A pre-trauma or never trauma-afflicted brain experiences events, creates thoughts and tells stories in a straight line. A happened, then B, then C. The past stays in the past, the present is the present, and the future unfolds step by step.

My brain does not work this way.

It does not experience time as a sequence. It experiences time as a network. A single moment does not live neatly in the past. It is always accessible, always active, always one wrong click away. I can look at something and be on a completely different plane within my own chronology, and have a trained understanding that I am also in the present and need to understand the two in order to not completely fall apart.

This is exactly how the internet retrieves information. There is no “past” or “present” online. A post from 2009 can be as immediate as a post from five seconds ago. A single search query can surface decades of history in an instant. The internet collapses time in the same way trauma collapses time.

A CPTSD brain does not process past events as over. The past is always running in the background, waiting for the right input to make it front and center again.

The same way a single keyword search can bring up years of digital archives, a single sensory trigger can bring up years of trauma. Both systems obliterate linear chronology.

Trauma memories are not recalled, they are relived. They do not stay confined to their chronological place in your life; they emerge suddenly, without warning, triggered by an association you did not even realise existed. It is completely, utterly unhinged and bizarre.

I love it for the way it makes me describe thing, like wine, like the world, how it processes the art and books and music that I love, but on the other side of that circle is the asme kind of intense and unpredictable, but coupled with the levelled frequency of dark.

This is also exactly how algorithmic feeds function. The infinite scroll never ends. The system is always scanning, always optimising, always categorising content to predict what comes next. You are never meant to reach the bottom of the page.

A CPTSD brain and an algorithm both assume that stopping is dangerous. Both believe that missing information is a potential threat. Both prioritise constant vigilance over rest. And in both cases, this is not a choice—it is a survival mechanism.

One of the most defining characteristics of trauma cognition that mimics an algorithm or predictive model in AI, is pattern-seeking.

The brain, having been hurt in ways it cannot afford to be hurt again, becomes obsessed with detecting signs of future harm.

It is constantly asking:

Have I seen this before? Does this situation resemble past danger? What micro-patterns can I detect that might indicate I am at risk?

It is doing this as fast as any AI we could possibly build. I know I am faster than ChatGPT in this way. The internet operates on the exact same premise. Machine learning algorithms function by recognising patterns in user behaviour, predicting outcomes based on past inputs, sorting information into meaning clusters.

A trauma brain and an AI system both operate on predictive models. Both are designed to assume that every input contains crucial, high-stakes meaning. A casual conversation might not be casual at all—it could be a test, a prelude to harm, a hidden repetition of something painful. Nothing is neutral, nothing is random. This is how a trauma brain stays alive. This is also how the internet keeps you engaged.

And in both cases, this is exhausting. I have to ask “what is happening” to myself after pretty much every interaction I could possibly have because I HAVE to go through all of the options before my brain will let me unclench.

I have never, ever been "entertained" by the internet, this "soothing" effect is because my mind is structured to accomodate spirals and fractures, all cross-referenced upon layers and layers of recursion. The internet is soothing to me, because to me it is a version of rest, one where my brain doesn’t have to accomodate the oppositional logic of the world outside of me.

It is a den of computational logic that my mind can sleep in.

My teenage years were spent inside the process of trauma rewiring—that slow, relentless restructuring of the brain in response to something it cannot fully process. It is beyond efficiently studied these reconnections your brain makes, and even now if I concentrate, I can feel the amount of electricity I generate to think the way that I do.

A trauma-rewired brain does not know how to stop thinking. It is constantly scanning, constantly assessing, constantly looping through every possible interpretation of a moment to prevent danger. Even in moments of supposed stillness, the mind is still processing, categorising, predicting.

This is why hypervigilance exists. Hypervigilance is not just an emotional state. It is a full cognitive reprogramming that tells the brain: never stop scanning for threats. Never stop indexing past experiences in case they are relevant. Never trust that what you see is the whole picture.

People with CPTSD often describe themselves as feeling like they do not have a single, stable self. Instead, they feel like they are a composite of different versions of themselves, activated by different contexts, a collection of conflicting identities, each shaped by different traumas, roles, and survival tactics.

A person who does not fully exist in a unified way.

This is precisely how online identity functions. This is precisely why the person I am on the Internet, in chat rooms, games, anywhere, there is simply for me no pressure to “make a me up”, because I feel made up as it is.

There is no singular self online.

There is only the self presented on social media, the self in private messages, the self shaped by algorithmic engagement, the self that exists as archived posts, ghost traces of previous opinions, versions of you frozen in time.

For me, I have no singular self in reality or the digital, and so I cannot experience what others do in the online world compared to the physical. They may as well be the same, they’re not, but it simply is not within my mental capacity to comprehend otherwise.

I cannot explain the feeling to you when I am playing something like Baldur’s Gate 3. I imagine it is the feeling that people have when they do not have a trauma-wired brain and they are out enjoying their life or sitting in the park with their friends. It is where I feel the most normal. It’s not to say I don’t get immeasurable and nourishing and life sustaining enjoyment from the park with friends, but it is this consistent, reliable and completely impenetrable sense of safety that I have within the framework of a games architecture and narrative that cannot be mimicked outside of a digital realm.

Where I can explain this a little further, is that a CPTSD brain and the internet both create identity through fragmentation. Both are built on the premise that there is no single, stable “I.”

If you have spent your entire life inside trauma rewiring, this feels normal. If you have spent your entire life online, this feels normal.

When I was a teenager, I used to stand in front of my mirror staring at my face, trying to recognise myself. I could not. I could not recognise myself. Still, when I recall people, I do not “see” them, but I see the patterns and the repetitions of them, and I pull them together to construct an image I can begin to trust. If you wonder why I have not recognised you, it is not because I am rude, it is simply because - I do not recognise you.

I have believed I am other people so completely, that I become them. I can make up stories so quickly and so strongly, because I am inhabiting them.

Where this becomes terrifying for me, is that I have learnt to live with this. Just.

Only recently am I in a place where I am happy, secure, gentle, kind. But I see people around me experiencing the things that I have my entire life, but in a way that is simulated, forced, not through re-wiring, but a forced conditioning or mimicry of the way my brain works and it honestly doesn’t surprise me the more and more people around me say they are “in a simulation” because if I was experiencing what they are, in the way they are experiencing what happened to me, I would never, ever of been able to get through it.

There is no such thing as a pre-internet mind anymore. There is no such thing as a generation that will grow up with a singular, linear sense of self. For them, there will be no difference between memory and archive, between cognition and algorithm, between self and feed. They are being raised inside a cognitive model that mimics trauma rewiring without ever having experienced trauma. You may call this age we’re living in one of unavoidable trauma - but that is false.

CPTSD affects around a maximum of 4% of the population globally. PTSD is an entirely separate phenomenon, and not to be confused. This affects around the same percentage of people.

What 100% of the population is this dissociation, depersonalisation and de-reality that is one of the symptoms of trauma-rewiring, but without their brains having the infrastructure to support it. It shouldn’t work, and it doesn’t, and that is what is happening to you, and everyone around you. It is impossible not to feel lost because of it.

This is to exist in a permanently untethered state. There is no rejecting of reality, because they simply have not experienced it in a way that we have studied it before. Wee have no idea what this will do to the future of human cognition.

I believe we need to start talking about the internet not as a medium, but as a neurological condition. It has already changed how we process time, structure memory, construct identity.

If you have CPTSD, you already know what this feels like. If you have spent too much time online, you already know what this feels like. The difference is—for us, this was a response to something.

For Gen Alpha, it is the starting condition.

People say Gen Alpha is worse. That they are the most nihilistic generation we have ever seen. I agree that something is happening. But it is not nihilism.

Nihilism suggests a conscious rejection of meaning, an active disengagement from reality. But what is happening here is not a choice. It is not defiance or apathy. It is something else entirely—something more fluid, more automatic, something embedded in the very structure of their cognition before they even had the opportunity to form an opinion about it.

Gen Alpha does not reject reality. They do not choose to disengage. They have never known a world that was not structured this way. Unlike previous generations who experienced the transition into a fully networked existence, Gen Alpha was born into it.

There is no "before" for them. No memory of a time when the physical world held primacy over the digital, when conversations weren’t immediately archived, when the feed didn’t feel like an extension of one’s own consciousness. They do not toggle between online and offline in the way we might perceive it. The separation has collapsed entirely.

Emotional intelligence, as most people understand it, is designed for a brain that was never forced to prioritise survival over connection.

For a trauma brain, the typical frameworks of emotional intelligence—openness, vulnerability, reciprocal trust—are nonsensical. They do not exist as instincts; they exist as threats. When being open once led to violation, when trust led to betrayal, your brain rewires itself to eliminate those risks entirely. So I had to train myself into a version of emotional intelligence that worked within the architecture of my mind, rather than against it. It was not intuitive; it was built, calibrated, and maintained through years of deliberate work.

Now, it is one of the most beautiful things about me—not a natural trait, but a discipline, an ongoing act of care for both myself and others. This is why drinking was so catastrophic for me: it wiped away that maintenance. It reset me back to the core functionality of my trauma brain, where there is no threshold for self-awareness, no space for emotional regulation—just raw, unchecked survival instincts.

And you see this across the board with trauma survivors: emotional intelligence is not an innate trait we lean on, but something we hold up with both hands, every single day.

For Gen Alpha, emotional intelligence will not be something inherited through natural socialization—it will have to be something actively taught, maintained, and contextualised within a world that does not organically reinforce it.

They are growing up in an environment where emotional responses are flattened into reactions, where selfhood is curated like a digital storefront, and where connection is mediated by engagement rather than presence.

The structures that once helped develop emotional intelligence—face-to-face interaction, unfiltered social feedback, the slow, messy process of learning emotional nuance through real-world trial and error—are being replaced by algorithmic validation, performance-based identity, and an overwhelming amount of emotional information with no clear framework for processing it.

If a trauma brain struggles with emotional intelligence because it was taught that vulnerability leads to harm, then Gen Alpha’s challenge is that they are being taught vulnerability is simply inefficient—that emotions are content, that identity is adaptable to the most rewarded form, that the line between sincerity and performance is irrelevant.

Their emotional intelligence will not be absent, but it will be structurally different—more externalised, more patterned, more influenced by mass feedback loops than by deep individual introspection. The question is not whether they will be emotionally intelligent, but what kind of emotional intelligence they will develop—and whether it will be something they can sustain, or something that sustains them.

They are the first generation raised entirely inside the cognitive architecture of the internet. Their reality has always been algorithmically mediated, always filtered through an ever-changing series of digital interfaces.

They do not "go online" because there is no such thing as "offline." The digital world does not sit parallel to their physical world—it overlays it, informs it, seeps into it until the distinction is meaningless. What we once called "real life" is now simply one part of a much larger, much more fluid existence.

Going offline would be the same as your partner going temporarily missing and then suddenly reappearing four days later, with simply no acknowledgement or explanation. That is the gravity of the feeling that Gen Z and Alpha are experiencing with their relationship to digital worlds.

For them, there is no bifurcation between the physical world and the online world. There is no “real” world. There is only the stream. Everything is content, everything is processed through the logic of virality, everything exists in fragments, stitched together by algorithms that dictate what is seen, what is engaged with, what is remembered.

And when all of experience is filtered through a system that prioritizes engagement over depth, performance over presence, spectacle over substance, the nature of perception itself begins to shift.

And this is why I push back against the idea that this is nihilism. Nihilism implies rejection, a conscious turning away from meaning, a decision to believe in nothing. But what is happening here is not rejection—it is dissociation. It is displacement. It is a fundamental rewiring of what meaning even is.

When meaning itself is algorithmically assigned, when emotions are flattened into reactions, when identity is something that must be curated and maintained like a digital storefront, the relationship between self and world begins to fragment in ways we are only beginning to understand.

Gen Alpha is not detached from reality because they refuse to engage with it. They are detached because they have never experienced it in the way previous generations did.

When they say their parents do not understand them, it is true, their parents do NOT understand them.

Their cognition was formed in an environment where reality is not a fixed, stable entity but something fluid, shaped by the logic of the feed. They do not remember a world before attention was a commodity, before every thought and action was shaped by an invisible audience, before existence itself became gamified.

Their minds were formed inside an infinite, liquid, always-on feed, a space where history collapses into an endless present, where context is optional, where the difference between reality and performance is no longer clear. And we have no idea what kind of cognition this is going to produce long-term.

It’s one of the biggest misunderstandings about this generation’s relationship with meaning. They aren’t rejecting meaning; they are floating in an environment where meaning is assigned externally, context is fluid, and there is no clear way to anchor anything.

Meaning isn’t gone—it’s just unstable. And that instability, more than anything else, is what creates the overwhelming, dissociative effect they are describing the only way they know how.

One of the most crucial shifts happening right now is the prioritisation of the external as the primary source of meaning. When identity, memory, and reality itself are structured by engagement, performance, and algorithmic validation, meaning is no longer something internally cultivated—it is something assigned externally, something proven through visibility, something that exists only insofar as it is witnessed and acknowledged.

The inability to recognise this shift, especially across generations, creates a growing chasm of misunderstanding. If you tell someone whose entire framework of selfhood has been shaped by digital existence that their world isn’t real, that their experiences don’t count, that meaning can only be derived from the physical, you are not pulling them back into reality—you are severing them from the only reality that has ever made sense to them.