And yet I don’t think the desire behind it is cynical. Community clearly matters to people in this space... perhaps more than almost anything else. When prices rise, community is everywhere. Holders celebrate together. Profile pictures proliferate. Discord is busy. People organise dinners, meetups, side events and group chats. They defend the project, make memes, buy more, recruit friends and speak sincerely about belonging.

Then number stops going up.

People disappear. The Discord quietens. The PFP changes. Conviction evaporates. The community that seemed so strong a few months earlier begins to look remarkably temporary.

For a long time, I thought this was mostly an indictment of speculation. People had confused financial alignment with cultural affinity. When everyone owned the same appreciating asset, of course it felt like community.

I no longer think that explanation goes far enough. Perhaps Web3 didn’t corrupt community. Perhaps it built exactly the form of association we should have expected from the people building it... and then called that community.

A Very WEIRD Kind of Belonging

I’ve been reading Joseph Henrich’s The WEIRDest People in the World. Early in the book, he asks a deceptively simple question:

Who are you?

Ask people to complete the sentence “I am…” repeatedly, and those from highly individualistic Western societies disproportionately describe themselves through their own characteristics:

I am curious.
I am ambitious.
I am creative.
I am a writer.

In societies organised more heavily around kinship, people are more likely to describe themselves relationally:

I am someone’s son.
I belong to this family.
I am from this village.
I am a member of this clan.

These aren’t simply different ways of introducing yourself. They suggest different ideas about what a person actually is.

Henrich’s larger argument concerns how Western, Educated, Industrialised, Rich and Democratic (WEIRD) populations became psychologically unusual. For what follows, one characteristic matters particularly: as intensive kinship structures weakened, Western societies became increasingly organised around autonomous individuals, impersonal institutions and voluntary associations.

Association became more portable. You could choose where to belong... and leave.

Crypto takes this psychology and gives it a price feed.

Control your wallet. Custody your assets. Choose a pseudonym, an identity, an affiliation, a community. And leave whenever you want.

There is something fascinating about a culture built so heavily around individual sovereignty subsequently becoming obsessed with the language of community. We built technologies designed to minimise dependence on other people and then began talking incessantly about belonging.

Maybe the contradiction isn’t accidental. Maybe one produced the other.

Financialising Belonging

Consider how strange an NFT community is as a social structure.

Take something resembling membership:

• Give it a price.

• Make that price continuously visible.

• Make membership globally tradeable.

• Allow anyone to enter if they can afford it.

• Allow anyone to leave whenever they can find a buyer.

• Create public rankings based partly on how many memberships people own and how desirable those memberships are.

• Give every existing member a financial incentive to increase demand for membership.

• Finally, call the resulting group a community.

This doesn’t mean the relationships that emerge are fake. Obviously they aren’t. Some of my own relationships in this space began through NFTs. People meet because of a shared asset and discover genuine intellectual, aesthetic or personal affinity. Relationships escape the collection that produced them. People collaborate, travel together, support one another and remain connected years later.

Something real happens.

But perhaps the token creates membership. Community is what may or may not happen afterwards.

Markets make the two particularly difficult to distinguish. Bull markets are terrible places to study community because almost every signal becomes contaminated. People attend the events because the events are exciting. They wear the PFP because the PFP confers status. They recruit new members while new members increase demand. They praise the founder while the founder is making them richer. They spend hours in Discord while their social life and financial interests have collapsed into the same place.

None of this means the relationships are insincere. It means sincerity is almost impossible to measure while incentives remain aligned.

Number go up creates a false positive for community.

A holder telling you how extraordinary their community is may believe every word. They also own an asset whose value rises if you believe them. Both things can be true.

This changes how I think about the NFT bear market.

We often talk about it as though it destroyed NFT communities. Perhaps it did something more useful: it tested them.

Remove appreciation.
Remove attention.
Remove status.
Remove the constant arrival of new members.
Remove the possibility that the JPEG in your wallet might change your life.

Now what?

The bear market didn’t merely make communities smaller. It removed many of the incentives that made membership look like community.

What remained became much more interesting.

The Exit Test

There is a simple thought experiment I keep returning to.

Imagine someone who owned a CryptoPunk for eight years. They participated in the culture. Made memes. Helped newcomers. Attended meetups. Wrote about Punks. Formed friendships. Contributed something meaningful to the history around them.

Then they sold.

What remains?

The blockchain has a wonderfully simple answer to the ownership question: their wallet no longer contains a Punk. But that tells us almost nothing about the community question.

Eight years of belonging cannot be erased by a transaction. Friendships don’t disappear when an address changes ownership. History doesn’t disappear because an NFT moves wallets. Contribution cannot be measured by the present contents of an address.

The chain is extremely good at answering one question:

What do you own now?

Community asks different questions:

What did you do?
Who did you meet?
What did you make?
What did you contribute?
What do you remember?
Who remembers you?
What remains?

This is the Exit Test.

The question is not:

Did you stay?

That would simply turn holding into another performance of loyalty. The better question is:

If you sell the token, what remains?

Someone can leave a genuine community. Someone can stop attending its gatherings. Someone can lose interest. Someone can sell everything. The point isn’t permanent participation.

The point is residue.

If nothing survives the transaction... no relationships, no memory, no continuing identification, no sense that something happened beyond ownership... perhaps what existed was closer to a club.

There is nothing wrong with clubs. But clubs and communities aren’t quite the same thing.

When the Token Stops Being the Point

Art Blocks makes the question more complicated. Its great speculative moment belonged to the NFT boom. The wider NFT audience that surrounded it contracted dramatically afterwards. Yet something around Art Blocks persisted.

Look at Marfa. According to an account of the event’s early years, a little over 200 people turned up in the first year. Around 400 came in the second. Around 700 came in the third. Art Blocks itself reported more than 700 attendees in 2023 and again in 2024. The gathering continued into a fifth year in 2025, with another planned for 2026.

That happened while the speculative NFT market was a shadow of what it had been.

More interesting still is how Art Blocks describes participation to artists:

“Show up over time, not just release time.”

Its guidance talks about sharing process, engaging with other creators and participating in conversations over time. It explicitly makes the point that the meaningful parts of releasing work can have little to do with selling out.

That sounds less like a marketplace talking and more like a culture discovering what survived its market. Somewhere along the way, the object around which people originally assembled became less important than the subject connecting them:

Generative art.
Creative coding.
Artists.
Collecting.
Aesthetics.
History.
Software.
And eventually, each other.

The token was scaffolding. Something escaped it.

Most token communities never made that transition. When attention and financial incentive disappeared, almost nothing remained. This may be one characteristic of the strongest token communities:

Eventually, they become communities about something other than the token.

CryptoPunks offers a different version of the same possibility. A Punk is an asset. But Punk is also history, language, aesthetics, lore, status, friendships, memes and memory.

People who no longer own one can remain part of that history. People who have never owned one can contribute meaningfully to its culture. The ownership boundary and the cultural boundary are no longer identical.

Something has leaked out of the token.

Residue.

The Right to Leave

The more I think about this, the less convinced I am that sovereignty is crypto’s deepest social value. There is a more precise word:

Exit.

Self-custody provides exit from custodians. Bitcoin provides exit from monetary authorities. DeFi provides exit from financial intermediaries. Permissionlessness provides exit from gatekeepers. Pseudonymity provides some degree of exit from inherited identity. Forking provides exit from consensus.

Nouns went so far as to build... and ultimately use... a mechanism through which dissenting holders could fork into a new DAO with a proportional share of eligible treasury assets. Faced with disagreement, the protocol guarantees exit rather than demanding reconciliation.

That may be excellent governance. It is also extremely crypto. And extremely WEIRD.

Crypto is extraordinarily interested in ensuring nobody can compel you to remain.

I don’t say this nostalgically. Communities that make exit difficult can be suffocating, coercive, exclusionary and cruel. Kinship can produce solidarity and nepotism. Durable relationships can protect people and trap them. There is no reason to romanticise a social order simply because its relationships are harder to escape.

The right to leave is important. But it produces a tension we rarely acknowledge. Crypto maximises the right to leave, while community derives much of its meaning from what survives that right.

I initially thought the missing ingredient was obligation. Perhaps community required us to stay when leaving became more advantageous. I’m less convinced of that now. Obligation implies that the person who stays is somehow more authentic than the person who leaves. Taken far enough, we simply recreate illiquidity as virtue.

There is another word I prefer...

Residue

Real communities leave something behind.

Not necessarily permanent membership. Residue.

Relationships acquire weight.
History accumulates.
People remember what others contributed.
You owe somebody a call.
You show up when attendance confers no status.
You explain the history to someone who wasn’t there.
You maintain the website.
You preserve the memes.
You defend an artist whose market has disappeared.
You organise the dinner.
You write the obituary.
You remember.

Something remains after the immediate reason for gathering has disappeared.

Community leaves residue.

And this is why “people who stayed through the bear market” isn’t a sufficient definition of community either.

Staying can be self-interested.
Staying can be stubbornness.
Staying can be bag-holding.
Staying can itself become status.

The person who sold may have contributed more to the culture than the person whose token never moved. The blockchain cannot tell the difference.

Which means none of us gets to use this argument too comfortably. Especially collectors. It would be very easy for those of us still here to divide the NFT world into two groups:

The mercenaries who came for number go up... and those of us who were here for the culture.

I don’t believe that.

Collectors rotate conviction too. We stop talking about artists whose markets have gone quiet. We discover new artists. We sell things that no longer fit our collections. We become excited when institutions validate artists we already own. We talk about historical importance while knowing exactly what the floor is.

I would like to believe I collect differently. That caring about the work rather than simply its price matters. That writing about artists, preserving their stories and continuing to collect through quiet periods constitutes something more durable than speculation.

Maybe it does.

But I also know how attention works. New things arrive. Markets move. Artists become fashionable. Others disappear from the timeline. Conviction has a remarkably convenient relationship with attention.

So I don’t think the interesting question is whether everyone else failed some purity test. The question is what any of us continue doing when nobody is watching.

Buying an artwork everybody wants is participation in a market. Continuing to care after the market has stopped caring begins to resemble something else.

I have increasingly come to think about collecting this way. Buying matters. It gives artists runway. But buying isn’t the entirety of patronage.

Writing, criticism, curation, advocacy and preservation give artists memory.

The same may be true of community. The transaction can bring us together. What happens afterwards determines whether anything remains.

The Permanent Object and the Disappearing People

There is an irony here that feels particularly Web3.

We became obsessed with permanence. The website might disappear. The marketplace might disappear. The company might disappear. The image itself might even disappear.

But the token persists. The provenance persists. The transactions persist. The chain remembers.

And yet many of the social structures we constructed around these supposedly permanent objects proved remarkably ephemeral.

The token stayed. The people left.

Perhaps we solved permanence for the easiest part: ownership.

The blockchain can remember that I bought something. It cannot remember why I cared. It can record that I sold it. It cannot determine whether I stopped caring. It can prove that two wallets once belonged to the same DAO. It cannot prove that the people behind them became friends. It can preserve a transaction perfectly while losing almost everything about what that transaction meant.

This is where another question begins: what can consensus remember, and what must instead be preserved by culture?

But it is enough here to recognise the boundary. The chain can preserve membership. People have to preserve community.

We spent years talking about putting communities on-chain. I don’t think we ever did.

We put membership on-chain. We put ownership on-chain. We put transactions, governance rights and provenance on-chain. And because these things were visible and measurable, perhaps we mistook them for the community itself.

But community seems to live somewhere the blockchain cannot easily see: in relationships that survive the sale; in work undertaken without expectation of return; in stories people continue telling after attention has moved elsewhere; in the things we preserve because losing them would matter; in some residual sense that these are still my people, even when saying so confers no obvious advantage.

Perhaps that is why the bear market mattered. It didn’t simply destroy Web3’s communities. It removed many of the incentives that made membership look like community.

What remained afterwards was considerably smaller... but perhaps considerably more real.

Crypto has spent fifteen years perfecting the right to exit.

Community might begin with what makes us stay.

SonOfLasG 🧙🏿‍♂️

@Sonoflasg

Using a machine that invents plausible memories to mourn the absence of real ones...

A dream of a dream.

We're used to AI art announcing itself through mutation. Malformed bodies, surreal combinations, visual excess. The uncanny. It says, look what the machine can imagine.

Wubbushi

@wubbushi

Part time scientist, full time artist

Check out my new collection
ExaltedLove.com

13

7:52 AM • Aug 7, 2026

The first thing that struck me about Wubbushi's HOLY was not that the images looked good.

AI images have looked “good” for a while now. They have been technically impressive, compositionally seductive, sometimes beautiful, sometimes grotesque, often instantly recognisable as machine-made. What felt different here was not quality in that narrow sense.

It was conviction...

The images behaved like photographs. Not illustrations, not surrealist composites, not obvious hallucinations. Photographs. The kind you find in a box after someone dies. I have been in Nigeria recently, sorting through my late father’s things, scanning old family pictures, and that may be part of why HOLY hit me so directly. These are the kinds of images you scan badly. The kind you look at for longer than expected because they seem to contain a version of someone’s life that was both ordinary and unreachable.

And yet the camera was never there. That is the tension at the centre of HOLY. This probably happened. This photograph never existed. That contradiction is not incidental to the work. It is the work.

HOLY presents itself as a reconstruction of a lost evangelical adolescence. Family albums have disappeared. The inherited camera survives as an object, but not as an intact archive. What remains are memories, dreams, emotional residues, fragments of religious childhood, images that should exist but do not. The model is asked to make the missing photographs.

That premise is important. Without it, HOLY could be dismissed as another exercise in synthetic nostalgia... fake grain, flash, motel pools, old America, religious iconography, thrift-store weirdness. The internet is full of images that look like that now. Prompt enough references into a model and it will give you something that feels like an old photograph.

But HOLY is not simply nostalgic for old photographs. It is about the absence of photographs. That is why the work lands differently. It is not only simulating a look. It is simulating a gap.

Family albums have always been unreliable. They leave out more than they preserve. They turn life into birthdays, holidays, Christmas mornings, smiling groups, acceptable evidence. They rarely keep the moment that actually formed you. The boredom before church. The threat beneath affection. The hot, cheap brightness of a motel pool. The social pressure of belief. The unphotographed confusion of being young inside a world that tells you love and prohibition are the same thing.

Nobody photographs the part you remember... HOLY tries to photograph that part after the fact.

This is where the work moves beyond technical novelty. The images are compelling because they seem to recover something that photography itself may never have captured. They do not document events. They document the feeling that events took place. They are not records. They are evidence-feelings.

That phrase resonates because it points to the new condition AI images are entering. For years, generative image-making was easiest to understand through failure. Extra fingers. Melted faces. Warped bodies. Impossible rooms. Torsos fused together. Surfaces that behaved like skin, plastic, fog and meat at the same time. The image announced its own synthetic condition. It did not let you forget the machine.

In some of the best early AI post-photography, that visible failure became the point.

rainisto’s Life in West America remains, to me, one of the clearest high points of that phase. Its America is recognisable but broken. It has diners, highways, domestic scenes, bodies, cars, faces, families, gestures we think we know. But everything is slightly or violently wrong. People seem assembled out of memory damage. Bodies mutate. Rooms sag into dream logic. The photographs feel like documentary images from a place that cannot exist.

That was the power of the work. It did not pass as photography. It used the failure to pass as its language.

Life in West America made the uncanniness of early AI feel culturally meaningful. The malformed body was not just a flaw. It became a sign of America being remembered through a machine that did not understand America, bodies, family, desire, loneliness or history, but could still generate the surface of all of them. The mistake was the message.

HOLY belongs to another moment.

Not because it is better. That would be too simple. Rainisto’s work may still be the stronger artistic achievement in the earlier AI post-photographic mode. But HOLY marks a shift in the image’s behaviour.

The obvious mutation recedes. The spectacle is suppressed. The image no longer says: look what the machine can imagine.

It says: this happened. Only the context tells us it did not.

That is a profound change. The early uncanny gave the viewer distance. However seductive the image became, the errors protected you. You could not fully believe in it. You were always brought back to the fact of generation.

HOLY removes some of that protection.

Its best images have the casual authority of vernacular photography. They look found rather than made. A boy in hard flash. A church exterior beside commercial signage. A tired domestic interior. A pool. A parking lot. A religious gathering. A body caught between innocence and performance. These images do not ask to be admired as AI. They ask to be believed as fragments.

That is why they produce nostalgia so effectively. Not because they reproduce the past accurately, but because they reproduce the way photographs taught us to feel about the past.

Photography has never been neutral memory. It is one of the things that trained memory. Most of us do not remember childhood separately from the visual forms that later organised childhood for us... family albums, home video, cinema, advertising, yearbooks, magazines, television, the half-remembered texture of cheap prints and badly stored negatives. We remember our lives, but we also remember the images that taught us how lives are supposed to look when remembered.

AI understands that second part very well.

It does not remember childhood. It does not remember church. It does not remember shame, belief, family, heat, boredom or adolescence. It has no father’s camera. It has no lost album. It has no private wound. What it has is the archive. And the archive has learned the look of memory.

That may be the most important difference between early AI post-photography and what HOLY points toward. The model is no longer merely producing impossible images. It is producing plausible cultural memory. It can now imitate not just the surface of photographs but the emotional function of photographs... the feeling that something was there before us, that a moment existed, that a life left a trace.

Photography’s old authority came from contact. However manipulated the final image became, something had stood before the lens. Light touched a body, a wall, a room, a face, a road, a wound. There was an event. The photograph could lie, but it lied around a point of contact with the world.

AI severs that contact.

No shutter opened. No light touched the subject. No child stood by the flag. No preacher stood under the sign. No exhausted boy was caught by flash in that exact way. And still the image arrives carrying the emotional weight of a record.

That is why “post-photography” is both an irritating term and a useful one.

Photographers are right to be suspicious of it. The phrase can sound like theft... synthetic images borrowing photography’s authority without accepting its discipline, labour or relation to the world. No one called photography “post-painting” simply because it changed the image. So why should camera-less images inherit the language of photography?

The objection is fair.

But the condition exists whether the term is elegant or not.

These images look like photographs, circulate like photographs, and trigger responses historically attached to photographs, despite having no photographic event behind them. They are not photographs in the old sense. But they are not simply illustrations either. They occupy the space where photographic belief has detached from photographic process.

HOLY is one of the first collections I have seen make that detachment feel emotionally convincing at scale.

That does not mean the work escapes the problems of AI image-making. It actually makes them more visible.

A few weeks after the collection began circulating, @sabatobox posted a side-by-side comparison between HOLY #130 and a lecture-hall still from the Coen Brothers’ A Serious Man, shot by Roger Deakins. Both images show a professor standing before a long blackboard covered in equations, students seated in rows, the same broad institutional seriousness, the same period atmosphere. The resemblance is hard to miss.

As a simple accusation, I find that less interesting.

Artists have always borrowed. Images have always echoed other images. Cinema borrows from photography. Photography borrows from painting. Advertising borrows from cinema. Family photographs borrow from advertising without knowing it. Memory itself is full of stolen compositions.

But as a symptom, the comparison is stark.

HOLY is a project about reconstructed memory. Yet in this case, what seems to surface is not only private memory. It is cinematic memory. The model, asked to produce a convincing room of thought, authority, education and period seriousness, appears to have reached toward an image already present in the wider visual archive.

That does not collapse the work... It complicates it in exactly the right place.

Because once AI images become photographically convincing, the question changes. We are no longer only asking whether the hand has six fingers or the face has melted. We are asking where the image’s conviction came from. What did the model absorb? Which visual histories are being reassembled? When does influence become leakage? When does cultural memory become recognisable source?

The stronger the image, the sharper the problem.

That is the uncomfortable bargain at the centre of HOLY. The same quality that makes the work moving also makes it unstable. The images feel like missing photographs because they have inherited the grammar of photographs. They feel like memories because they are built from the debris of other people’s memories.

This is not a reason to dismiss the collection. It may be the reason to take it seriously.

The fantasy of pure originality was never very useful, but AI makes it impossible to maintain. These systems do not create from nowhere. They create from the sediment of the visual world: family albums, fashion spreads, film stills, documentary images, religious pamphlets, youth photography, stock images, scanned prints, anonymous uploads, famous compositions, forgotten photographs, everything that has been fed into the machine directly or indirectly.

But human memory is not pure either.

We remember through contamination. Through films we watched too young. Through adverts. Through album covers. Through family stories. Through photographs we saw so often they replaced the event. Through images of places we never visited but somehow feel we know. A person remembering their childhood may also be remembering the way childhood has been pictured back to them.

HOLY sits inside that contamination rather than outside it.

That is why it feels contemporary. It does not offer the past as something recoverable. It offers the past as something reconstructed from loss, image culture and desire. It is intimate and derivative at the same time. It is personal and impersonal. It mourns an archive by using an archive.

That is the part I find difficult to shake.

Using a machine that invents plausible memories to mourn the absence of real ones sounds, at first, almost perverse. But it may also be brutally accurate. What else would a lost family album become now? Not silence. Not only grief. A prompt. A reconstruction. A darkroom with no negative. A dream of a dream.

The question is not whether the result is true... It is not. The question is whether it can touch something true... HOLY suggests that it can.

Not always. Not cleanly. Not without debt. Not without the risk of source images coming through too clearly. Not without borrowing the emotional authority of photography while removing photography’s contact with the world.

But sometimes the false photograph gets close to a real absence.

That is why I feel this collection is important. It does not prove that AI has surpassed photography. It does not make cameras obsolete. It does not settle the argument about post-photography, authorship, fair use or the ethics of training data.

It marks a perceptual shift.

Before, AI post-photography was often compelling because it could not fully pass. Its failures kept the machine visible. The uncanny was the signature.

Now, with HOLY, the image can begin to pass emotionally even when we know intellectually that it is false. The work does not ask us to marvel at mutation. It asks us to confront conviction.

After uncanny, distortion is no longer the central problem.

The problem is belief.

And belief has always been the most dangerous thing images do.

The more interesting story is not about money or JPEGs but about consensus. Money was simply the first thing we discovered decentralised consensus could secure. Everything that followed has been an attempt... sometimes deliberate, more often opportunistic... to attach durable claims to things that are harder to measure... cultural reference, participation, status, belonging.

Money has always depended on collective agreement. A pound is valuable because enough people agree that it is. What Bitcoin demonstrated was that this agreement no longer required a central authority. A distributed network could maintain a shared ledger of ownership with nobody in charge of it. That achievement is usually understood as the creation of digital money. It was also something broader... the discovery that decentralised consensus could create durable, shared memory. Once that became possible, a more interesting question followed.

What else can consensus remember?

Ethereum widened the scope. If consensus could maintain ownership of money, perhaps it could maintain ownership of any digitally native asset. The first generation of NFTs (CryptoPunks, CryptoKitties, Autoglyphs) explored that proposition as a technical matter, each demonstrating that unique digital objects could exist independently of the platforms that hosted them.

Today we think of CryptoPunks as cultural icons. At launch they were experiments. The culture, and the capital, came afterwards.

Others were asking a different question. Not whether digital objects could be scarce, but which digital objects were worth making scarce. A subtle distinction, but it separates two lines of inquiry. One explores the mechanics of ownership. The other explores the objects of ownership.

Rare Pepes were among the first serious attempts at an answer. From 2016, on Counterparty, collectors issued scarce digital cards based on Matt Furie’s Pepe the Frog. They were not attempting to own Pepe. The character remained infinitely copyable, endlessly remixable and culturally alive well beyond the blockchain.

What they created instead was a scarce, persistent claim attached to that cultural phenomenon. One the surrounding culture could choose to recognise as canonical.

The meme remained free. The claim became scarce.

CryptoDickbutts arrived at the same idea from a different starting point. Dickbutt already belonged to the internet. Drawn by K.C. Green in 2006, the character escaped its original comic and spread through forums, imageboards and private chats until it became part of the internet’s shared vocabulary.

CryptoDickbutts did not put Dickbutt onchain. It created a durable claim associated with him. Not ownership of the character, not ownership of the joke, but a socially recognised position connected to both.

Participation became ownable. The culture did not.

This is where NFTs are most often misunderstood. People ask why anyone would pay for an image that anyone else can copy, but the question assumes the image is the asset, and usually it isn’t.

The image is an anchor. The asset is a socially recognised claim whose record is maintained by decentralised consensus and whose meaning is maintained by culture, and by capital.

A blockchain does not prevent copying. It records the claim. The surrounding culture determines whether that claim continues to be recognised.

Machine consensus preserves the record. Social and economic consensus decide whether it matters.

Memecoins explore the same territory through fungibility rather than uniqueness. Dogecoin turned an internet joke into money. PEPE turned an internet character into a financial network. dogwifhat reduced the idea to its limit. A photograph of a dog in a knitted hat became the basis of a market because enough people collectively decided that it mattered.

These are easily dismissed as speculative manias, and sometimes that is exactly what they are. But speculation is not merely noise layered on top of culture. It is frequently the mechanism through which recognition is tested and amplified. Markets are one of the ways communities discover whether a cultural claim is durable enough to persist, and whether it is worth coordinating around. Most fail. A handful survive, usually because capital and attention continue to reinforce them long enough for social recognition to harden.

NFTs and memecoins are usually presented as competing asset classes. They look closer than that. NFTs create persistent claims on unique cultural objects; memecoins create persistent claims on shared cultural narratives. One emphasises provenance, the other participation. Both rest on the same underlying mechanism: consensus.

But again, the work is divided. The protocol establishes who holds the claim. The culture and capital decide what holding it means.

None of which means blockchains own culture. Culture remains open, contested and infinitely reproducible. Dickbutt still belongs to the internet. Pepe still belongs to the internet. People continue to copy, reinterpret and transform them without reference to any ledger.

What blockchains create are durable claims that communities... often overlapping with token holding communities... may choose to recognise as canonical. It can also disappear when the capital leaves or the joke grows stale.

The ledger preserves the claim. Only culture decides whether the claim continues to matter.

Seen this way, blockchains are becoming a new kind of infrastructure. Not merely financial infrastructure, and not simply computing infrastructure.

Memory infrastructure... although only in a narrow sense.

A blockchain can preserve provenance, sequence and ownership. It can record the claims around which a culture organised itself. It cannot preserve the lived experience that made those claims meaningful.

The chain remembers the coordinate. The culture remembers why it mattered.

For decades the internet has produced culture faster than any civilisation in history... memes, communities, shared references, digital folklore, status symbols. It became extraordinarily good at generating meaning, but had no native mechanism for preserving stable claims around that meaning.

It could preserve the image while losing the provenance. Preserve the joke while forgetting its route through history. Preserve the culture while offering no durable way to locate a claim within it.

Blockchains changed that. Not by deciding which claims deserve to endure, since cultural consensus remains selective, contested, often captured by capital, and frequently wrong, but by giving communities a durable ledger on which claims can be recorded, challenged, traded and, occasionally, remembered.

Most of those claims will not last. Most NFT collections will disappear and most memecoins will be forgotten. That should not surprise us, because most cultural artefacts disappear too.

The interesting question was never why so many fail. It is why a few continue to command recognition long after the speculation has faded.

We tend to describe crypto as moving value onchain. Perhaps the sequence runs in the opposite direction: consensus comes first, and value follows.

Bitcoin showed that machine consensus could secure money. Rare Pepes suggested that persistent claims on internet culture could be recorded on the same rails and recognised by a community as canonical. CryptoDickbutts extended the idea. Memecoins demonstrated that shared narratives could themselves become economically legible.

Whether identity, reputation and other forms of social capital can follow in a similarly durable way remains an open question. The infrastructure is there. Whether culture... and capital... will choose to use it for anything lasting is another matter.

Money was never the destination. It was simply the first thing consensus learned to remember.

Every blockchain develops its own artistic culture. Spend enough time moving between Ethereum, Tezos, Solana and Bitcoin and the differences become obvious. Different aesthetics emerge. Different artists thrive. Different collectors are drawn in.

But the more interesting difference isn't the art... It's the role the ecosystem asks its collectors to play. We spend a great deal of time discussing what should be collected. Far less time asking how one ought to collect... or why the answer changes from one chain to the next.

I've argued before that a blockchain is a petri dish. A cultural medium with its own conditions, in which certain things grow and others cannot. What I under-stated in that essay is where the collector sits in the experiment.

The collector is not the observer at the microscope... The collector is in the dish.

Nutrient. Selection pressure. Environmental condition. What collectors reward becomes what the culture produces. What they ignore, dies. Over time, those behaviours harden into the character of the chain itself.

So the question is not which chain has the best art... The question is: what does this ecosystem need from its collectors in order to survive?

The role is set by what the chain lacks

The mechanism made plain.

Every ecosystem has a scarcest resource. The collector's role is determined by that scarcity. Not by ideology. Not by community values. By what is missing.

Ethereum lacks nothing except attention. It has the canon, the capital, the institutions, the auction results, the museum acquisitions. What it cannot manufacture is legitimate discernment. Ten thousand collections launch; three matter. The Ethereum collector's work is telling them apart. The role is curator. Every serious acquisition on Ethereum is an argument about what belongs in the canon. The argument is expensive. It spends reputation as well as ETH, and only one of those can be replenished. The collector who cannot make the argument is merely buying.

Tezos lacks capital. Its culture is rich and its wallets are thin. The distance between a sale and no sale is the distance between an artist continuing and an artist stopping. Fifteen tez does something there that 0.3 ETH does nowhere. The role is patron. On Tezos you are not selecting from an abundance. You are funding a practice into existence. The object you receive is almost incidental. What you are actually buying is the next body of work... the one that now gets made.

Bitcoin lacks a past. Or rather... it has the deepest past in the ecosystem and the shallowest cultural memory of it. A chain obsessed with permanence, still assembling the story of what it has permanently recorded. The role is archivist. But not the archivist of the quiet reading room. The significance of these artefacts is not unsettled; it is actively contested... often by the chain's own culture. Maximalism denies that Bitcoin is an art medium at all. Parts of ordinals culture are aggressively anti-narrative. The archivist here is not waiting for consensus to arrive. They are preserving evidence against resistance. The collecting is a wager that history will need witnesses... placed in a room that keeps insisting there is nothing to witness.

The Solana avant-gay lacks institutions entirely. No canon, no galleries that matter, no settled market. The scene exists only in the ongoing activity of the people inside it. What is scarce is liveness itself. The role is collaborator. Stop participating and there is nothing to collect, because the work is the participation. Non-collaborative collecting exists there, and the scene reads it precisely: you buy, you go silent, and the purchase registers as absence. The money arrived. The collector never did.

Notice that none of these is a description of a blockchain.

Each is a description of a job.

And jobs change hands. The role is indexed to the chain's present scarcity, not to its essence. Ethereum asked for patrons in 2019... that is what the early Art Blocks collectors were, funding a practice nobody had yet thought to call a canon. Then the canon arrived, capital followed, and the ask changed. Tezos's patronage may be a stage rather than a nature. Solana's liveness problem lasts exactly as long as its institutions fail to appear. The dish is not fixed. Its chemistry drifts, and the role drifts with it.

This makes the determinism stronger, not weaker. A stereotype stays still. A mechanism updates.

The objection

The obvious objection: collectors move.

And it's true. Artists are sessile. A practice sinks into its chain... the tooling, the audience, the provenance of an entire body of work. When a generative artist builds on fxhash or Art Blocks, the chain is not a venue. It is inside the work. Institutions are more rooted still; a platform is its chain.

Collectors are migratory. A wallet crosses chains in an afternoon. The same collector can hold Ethereum blue chips, fund Tezos editions and sit inside a Solana scene without any of those ecosystems remaking them.

So the determinism seems to fail at exactly the point where the essay needs it.

It doesn't. It relocates.

Mobility does not exempt the collector from the role. It only settles who owns it: the chain, not the person. When you collect on Tezos, you are doing patronage whether you intend it or not, because that is what money means there. When you spend a week weighing traits across a grail collection on Ethereum, you are doing curation, because on Ethereum that is what discernment consists of.

The role attaches to the act, not the person.

What mobility buys the collector is code-switching. The multilingual speaker changes register with each language without ceasing to be themselves. The collector changes obligations with each chain without ceasing to be one collector.

The freedom lives between the dishes.

The determinism holds within them.

This is also why artists who work across chains fascinate me. Some dabble experimentally; some move with great intention. Either way, the chain shows up in the work... the register shifts, the address to the audience shifts, the edition logic shifts. An artist cannot visit a chain lightly. The medium gets into the output.

When a collector moves chains, the chain shows up in the conduct instead.

The artist is shaped deeply and slowly.

The collector is shaped shallowly and fast.

Both are being shaped.

Against the grain

Three harder cases.

First: the collector who refuses the assignment. The pure speculator on Tezos. The pure patron on Ethereum. What happens to them?

Nothing. That is the point.

The environment does not forbid deviant strategies. It declines to compound them. The speculator on Tezos is not punished; no status accrues, no standing builds, and the strategy simply fails to grow. Selection pressure rarely looks like punishment. It looks like being ignored.

But that is true only of individuals.

Second: speculation at scale. High-velocity financialisation is not a strategy inside the environment. It is weather. It does not sit beside the art scene; it irradiates the dish. And this is not an exception to the mechanism. It is the mechanism's origin... though the radiation works differently on each chain. On Solana it is chronic: a constant background dose that dissolves institutions before they can form. Liveness is scarce because nothing is allowed to settle. On Bitcoin it was episodic: one great wave, a smaller one after, and then the attention left with the price action. The inscriptions remain; the crowd does not. Memory is scarce there not because artefacts arrive too fast to narrate, but because almost nobody stayed to narrate them. The archivist works the quiet after the boom... which is, historically, when archives have always been built. Tezos is the mirror case: the radiation never arrived, and so capital is what is missing. The mutagen sets the scarcity. The scarcity sets the role... sometimes by its presence, sometimes by its passing, sometimes by its absence.

Nor is the radiation simply damage. Each boom leaves a stratum. The archivist's material is speculative sediment; ordinals exists at all because of speculative energy. Mutation destroys cultures and creates them, indifferent to which.

Third: capital. A well-capitalised collector can appear to override the assignment entirely. Arriving on Tezos with Ethereum money and Ethereum habits, curating where the chain asks for patronage. Doesn't power break the model?

No. A large enough organism does not escape the medium. It becomes medium.

Deploy enough capital into a capital-poor ecosystem and you change what the ecosystem lacks... and therefore what it asks of everyone who arrives after you. Power does not exempt you from the dish. It promotes you from participant to environmental condition.

The whale is simply the mutagen with a name.

Residence time

One caveat, because the mobility argument can be pushed too far.

Duration converts role into character.

The collector who stays put long enough stops performing the role and starts being formed by it. Three years on ordinals and you genuinely think like an archivist. It is no longer a register you switch out of. Fund fifty Tezos artists at five tez each and you have not merely behaved like a patron. You have acquired the sensibility of one.

The dish gets the collector too. It just needs residence time.

So the framework has three speeds. Institutions are shaped fastest and most completely. Artists are shaped deeply, through the work. Collectors are shaped last... and only through commitment.

Which gives the collector something neither artists nor institutions have: the choice of how deep to go.

Three confessions

A framework is easier to build than to stand inside. So let me test this one on myself.

I hesitate on Tezos. I love a great deal of the work there... and that is precisely the problem. I can read the contract before I sign it. Patronage is not a purchase; it is an undertaking. It asks for continuation. For keeping up. For the next edition, and the one after that. I feel that weight before the wallet opens, and it often stops me. I used to read the hesitation as indecision. I now think it is comprehension. The obligation arrives before the signature... and this essay began there, in a week of reading contracts I did not sign.

On Solana I collect and remain an outsider. Nobody imposes this. The scene is not gated; it is simply earned. Which highlights a distinction the taxonomy should make explicit: the collaborator is the only role that cannot be bought. Curation can be performed with capital and research. Patronage with capital and constancy. Archiving with capital and conviction. Collaboration is paid for only in presence, and presence has no floor price.

Ethereum is where I am most comfortable. The curator's work, investigating, contextualising, building the case before pursuing the piece, suits my temperament. Or so I would like to think. But my own argument will not let me say it so cleanly. Ethereum is where I have lived longest. The comfort may be evidence that the role fits me. Or it may be evidence that the dish has already done its work, and formed the temperament that now feels like a fit.

I cannot tell from the inside. Neither can you.

Obligations

We describe NFTs as changing ownership.

Increasingly I think they change obligations.

A collector on Ethereum, Tezos, Bitcoin and Solana may spend the same money. They are not signing the same contract. One is preserving a canon. One is funding a practice. One is safeguarding a history. One is keeping a culture alive by performing it.

And the contracts do not travel. Bridge your assets and your obligations stay behind; new ones are waiting on the other side. You do not sign once, as a collector. You sign again every time you cross.

This is, I think, what collecting onchain actually is, underneath the acquisitions. You do not choose your role. The chain assigns it. What you choose is the dish. Which scarcities you will answer to. Which selection pressures you will stand inside, and for how long. Choose briefly and you code-switch. Stay, and the medium begins its work on you.

So the question worth asking of a collection is not whether the right pieces were bought. It is: what is this making me? Not what kind of collector I am. What kind I am becoming.

He put it plainly on X in July 2026: your past forms your present, but it might not be your past. Most artist statements are decoration. This one is a mechanism, and Series 0 of Identity Studies is that mechanism running nine times.

The closed loop

The method first, because in AI art the method is where the ethics live.

Soria-Rodríguez hand-codes generative systems. He has done so for years... contrapuntos, entretiempos, ecologías, the abstract bodies of work released through fxhash and Art Blocks. For Identity Studies he trained bespoke image models on those systems' outputs, folded together with his personal archive: drawings of his daughters, photographs of his ancestors, scenes from his own life. The models generate only within the visual grammar this material teaches them. Nothing scraped. Nothing rented. No foundation model laundering a million strangers' pictures into a style.

Important, but not new. Helena Sarin was training networks exclusively on her own drawings in 2018. Anna Ridler photographed ten thousand tulips by hand and exhibited the dataset as the artwork. Herndon and Dryhurst built a public vocabulary for identity held in training data. The closed loop is a tradition, not an invention, and Identity Studies joins it rather than founds it.

What Soria-Rodríguez adds is a specific circuit the tradition had not yet wired: his own generative systems as one training substrate, his ancestral archive as the other, and identity itself (not style, not species, not voice) as the declared subject. He fed a machine the total recorded evidence of one self, then watched what kind of self came out.

The images are the experimental result. Every picture in Series 0 is his identity operating through a body that is not his. If the work feels like him, then identity has been demonstrated... not argued... to be a pattern that survives transfer between substrates. The medium is the thesis.

Rooms made of static

The pictures share one formal decision that separates them from the AI-figurative genre they superficially resemble.

The grain is not on the image. The grain is the image. Walls, floors, air... every room in Series 0 is a field of particulate noise, and the figures read as condensations out of that field. In Nº 03, a body reaches upward and its arm breaks apart at full extension, particles returning to the wall behind it. Garments lose their edges. Matter precipitates from static and dissolves back.

Call this what it is: the latent space made visible as architecture. Rainisto's grain is nostalgic... film stock, memory of a medium. Soria-Rodríguez's grain is ontological... the undifferentiated substrate before any self exists, out of which a self briefly holds shape. The series title is not poetry. A swirling shadow may transcend into an elephant or a dancer is the physics of these pictures, stated exactly.

And the quiet radical move... the rooms are made of the same noise as the figures. Container and contained are one substance. There is no self separate from its conditions. You do not stand in the room. You are the room, locally organised.

The procession

Then the model did something the artist did not caption, and it is the most important thing in the series.

Nº 02: a crowd gathered beneath a yellow-and-white canopy, tall poles raised, bodies pressed together in formation. Elsewhere, a tiny solitary figure in what is unmistakably penitential costume. In Nº 01, a cross and liturgical objects mounted on a bare wall above a collapsed body.

This is Semana Santa. Spanish Catholic processional memory, surfacing through a model trained on an ancestral photographic archive... unasked for, unprompted, sedimented in the data and rising to the surface the moment the data was made to dream.

Marianne Hirsch gave this a name in another context: postmemory. The memory of experiences you did not have, transmitted so deeply through family and ritual that it functions as your own. You did not walk in the procession. Your grandmother did. The procession walks in you anyway.

The model is the limit case of postmemory. Its entire past belongs to someone else. It produces a present from a wholly borrowed history... and in doing so it shows Soria-Rodríguez, from the outside, exactly what was inherited in him. You cannot see your own conditioning from within it. A machine trained on your traces can hold it up like a telescope holds up distant light.

This is the honest answer to the question every AI-art sceptic asks: what can the model show that the artist could not make alone? Here, precisely this. The inherited layer of a self, made visible to the self that carries it.

The wound dancer

Nearly every figure in Series 0 is mid-motion. Reaching, spinning, falling, caught between positions. Movement is the one thing in these sealed rooms the figures author... identity as verb, the self as something done rather than owned. The blur is the argument: no figure is captured at rest, because the self does not exist at rest.

Then Nº 05 complicates it. A ballerina spins on a small transparent disc... a music-box mechanism. The dance, the one apparently free act in the whole series, may itself be wound by someone else. Agency inside inheritance, possibly indistinguishable from it.

His caption for Nº 03 asks the question outright: are you in control, or led by an identity that lures you in? Lures. Identity as an agent with its own agenda... something that finds you, in the language of the project site, rather than something you build. The work declines to resolve this, and the refusal is what keeps it art rather than illustration.

The flags tweet supplies the political edge, understated but present: we go to war and shelter behind flags for our identity... yet how much of it actually comes from within? People kill and die for the inherited costume. The procession piece is that sentence, pictured.

Where it actually sits

Now the part a writeup owes its reader...

Identity Studies is not a founding work. The dataset-as-medium tradition has priority elsewhere... Sarin, Ridler, Tau, the Herndon–Dryhurst apparatus. It has no institutional record yet: no museum, no critical literature, a small direct market, nine 1/1s at 0.33 ETH from a project site. The philosophical territory... constructed selfhood, inherited identity, the self as pattern... is Hume's, Halbwachs's, Hirsch's. The work embodies these ideas with unusual clarity. It advances none of them. Embodiment is art's job; the distinction still deserves stating.

And the project's sharpest thinking lives in the wrong place. The site text gestures... understanding identity is essential for survival in a world with new intelligences... where the tweets cut. Your past forms your present, but it might not be your past is a better artist statement than the artist's statement. The real exegesis of Series 0 is a caption thread on a platform where texts go to disappear. Someone should archive it now, while it is complete and attributable. Consider this essay a start.

Marcelo Soria-Rodriguez

@msoriaro

your memory might just not be your memory.

which is not an empty statement, but rather loaded with truth. Your past forms your present, but it might not be your past.

17

10:05 PM • Jul 17, 2026

What the work has, and what most of the field lacks, is total alignment. Method, surface, and subject say one thing. The closed training loop enacts inherited identity. The noise rooms picture it. The procession proves the archive carries it. In a genre where hazy figures dissolve in empty rooms because the model's defaults tend that way, here the dissolution is load-bearing. That alignment puts Series 0 at the top of the artisanal band of AI-native work... below the canon-in-formation, above nearly everything else, and more coherent than several practices ranked above it on record alone.

The hinge

Series 0 is the visible pivot in a practice. Years of abstract generative systems, then the 2025 emotional-superposition series with Heft, and now this: the turn to the figure, the archive, the self. If the coming series consolidate (if the insight becomes apparatus, texts, a resolved body of work) this is the origin point people return to, and the writeups will say a hinge was acquirable for five hundred euros in the summer of 2026. If the practice stays in perpetual openness (the site promises nothing here is fixed) the hinge argument weakens, because a hinge needs a door on both sides.

That is the open question, and it is his to answer.

What is already answered is smaller and stranger and enough. A man fed a machine everything recorded of himself, and the machine returned his grandparents' liturgy. The past that formed him was never his. It formed him anyway.

Not your memory. Still your self.

That is the sort of artist he is. The interest sits in the process, and the image is the residue.

InkField reverses the apparatus without changing the conviction. Polypaths grows forward from initial conditions toward outcomes nobody chose. InkField works backward from an act that already happened: fifty painting sessions, each recorded not as a picture but as motion. The path of the brush, its speed, its pressure, its turns, and the pauses where thought interrupted movement. Each work is a self-contained package of code and data that re-performs the session live in the browser, stroke by stroke, whenever it is opened.

The distinction Wang draws is exact. This is not, he says, an attempt at a calligraphy-like brush. It is an attempt to turn the movement of the brush into something that can be recorded. Every stroke in traditional writing contains a start, a flow, and an ending. How intention enters, how force travels, how motion resolves. And all of it, he notes, usually lives only in the artist's body. What remains is just the result.

That sentence is the whole argument, and it is not an argument about artificial intelligence. It is much older than that. Every painting ever made has thrown away almost everything that went into it. The hesitation before a stroke, the acceleration through it, the moment of pressure where confidence arrives or fails... the finished surface preserves none of this. It preserves the outcome, and the outcome is the smallest part. We have accepted this loss for so long that we mistook the residue for the work.

One source

To understand what Wang is recovering, you have to go back before the separation that Western art takes for granted.

書畫同源. Calligraphy and painting, one source. Wang has explained it plainly: in the East, writing and painting were never two separate things. The same brush, the same paper, the same breath in that moment. When a calligrapher writes, each stroke is a word, and also the person making it. When a painter paints, each stroke is an image, and also the state of mind that produced it. The eight strokes hidden in a single character were used to teach both disciplines at once: learn those eight and you can write anything, and you can also paint anything.

I mention this because it corrected me. My first reading of InkField's most calligraphic works was that they were writing with the meaning burned off. Language reduced to its gestural residue... A Western instinct, and it is wrong. It assumes there was a semantic layer sitting on top of the gesture, waiting to be subtracted. There was never a layer. The stroke was always both things at once, and the tradition never asked it to choose.

Which reframes what Wang has actually done. He has not stripped writing down to movement. He has built a medium in which the original unity (stroke as image, stroke as person, stroke as the state of mind behind it) can be preserved rather than inferred. The tradition always claimed that the mark carried the maker. The tradition could never prove it, because all it could hand you was the mark. InkField hands you the making.

The score and the performer

Here is the mechanism, and it is more subtle than it first appears.

The stroke data is deterministic. Path, timing, sequence, pressure. The same gestures in the same order, on every viewing, on any device. That layer is archived. It does not drift.

The ink is not. The diffusion, the bleeding, the dry-brush breaks, the way a wet edge spreads into the paper: these are simulated fresh each time the work runs, with small variations deliberately built in. Open the same work twice and the hand will have done the identical thing, and the ink will have answered slightly differently.

So the work is a score, and the score is the hand. What performs it is the ink.

This is not a technical curiosity. It is a precise model of the only thing in painting that was ever unrepeatable. A calligrapher who writes the same character twice makes the same movement and gets a different mark, because the paper is not a party to the intention. Absorbency, ink load, humidity, the grain of the sheet... the material has its own opinions and expresses them at the moment of contact. Wang has not simulated painting. He has simulated the relationship between intention and material, which is the only part of painting that could never be repeated and therefore the only part that was ever alive.

He is careful about how far he pushes this. The JSON file, he says, is not an image but a continuous state of motion... and then he hedges, precisely: if this data has value, it may be closer to human thought than the final form itself. More about how something is made than what is made. I would leave the hedge in place. The tentative version of that claim is more interesting than the confident one, and more likely to be true.

Number 1

Number 1 is where the strokes come closest to writing and refuse to arrive. They gather into columns that approach characters and stop short, black massing into dense knots down the vertical bands of the square. It is the opening position of a hand-curated fifty, and it is a statement of the collection's premise: this is what the eight strokes look like when they are asked to be nothing but themselves.

Watch it perform, though, and the finished surface turns out to be the least of it. While the session replays, the work shows its own skeleton. The recorded path appears as thin red vectors running ahead of the ink (the data drawn as data). A ruled grid materialises around the column being written, the same armature as the gridded practice sheets on which every calligraphy student learns proportion. For the duration of the performance, intention, training, and act are visible as three separate layers. Then the last stroke lands, and the first two withdraw. The finished image keeps nothing but the ink.

This is the thesis staged as theatre. Wang's argument is that the result conceals everything that made it. Number 1 does not merely illustrate the claim... it performs the concealment in front of you, every time, and completion is precisely the moment the process disappears. The ordinary condition of painting, the loss of everything except the outcome, is re-enacted at the end of every viewing. The difference is that here you were shown what is lost before it went, and tomorrow it comes back.

It turns out the work held its position for the artist before it held it for anyone else. Wang has said that while testing the system in a museum setting he kept returning to Number 1 as his reference (the piece against which the whole apparatus was judged) and that it carries the heaviest structural weight in the series. In a catalogue of individually made works, the artist's own benchmark is not one output among fifty. It is the work the system had to satisfy.

Not a message withheld. The moment before a stroke has to decide what it is.

Number 34

Number 34 is a handscroll, and it is one of a few in the fifty that understands what a handscroll is for.

The handscroll is the one format in art history built explicitly to be read in time. You cannot see it at once. You unroll it, section by section, and the composition arrives as a sequence... it has an opening, a development, a dissolution. It was the first solution to a hard problem: how does a static medium hold duration? The answer was to make the viewer supply the time.

Number 34 obeys that grammar rather than merely borrowing its proportions. The gesture mass gathers at the left, reaches full density through the centre where threads of gold concentrate, and breaks apart toward the right. Beginning, journey, ending. Read across, it behaves like a scroll and not like a wide picture.

And in this medium the format has stopped being a workaround. Duration is now native to the work... every InkField re-performs itself in time whether it is square or panoramic. So the scroll no longer has to impose time. It keeps it. The oldest solution to the problem and the newest one, holding the same object.

What the chain is for

Everything a painter knows in the body is normally lost. That is the ordinary condition of art, and it is the condition InkField declines.

We have spent several years asking blockchains to remember the easy things. Ownership, scarcity, sequence... and above all the image, hashed and pinned and called permanent, as though the picture were the achievement. It never was. The picture is what was left over. Freezing it forever is an act of extraordinary confidence in the least important layer.

Wang does not freeze the image. He fixes the decision... the hand, the pause, the timing of pressure coming down... and lets the substance stay mortal, re-improvised on every viewing, gone by morning, made again tomorrow slightly otherwise. Permanence for what was intended. Impermanence for what it was made of.

An image can be imitated, he writes. But the hesitation, the pause, the timing of a hand pressing down... that only belongs to the person it came from.

That is the right thing to make permanent, and it is the first time we have had a way to do it.

Aluan Wang, InkField, fifty works, 2026. Numbers 1 and 34 are in my collection.

I had been scrolling backwards through a man's timeline for a week.

Not for a reason. Not at first. I had been holding a Dragon since 2021 and I went looking for the beginning of it, the way you sometimes go looking for a photograph of a friend before you knew them. What I expected was juvenilia. Early code, rougher, recognisably him.

What I found were rugs. Or what I began thinking of as the rugs (he never called them that). A run of early experiments: long striations in ochre and rust and cobalt, packed to the edge of the frame, dense as a woven thing. Some read as carpet. Some read as loom. All of them read as touch... as if the file had a nap to it, as if you could run your palm across the screen and feel the direction of it. They sit in his timeline before anything anyone knows him for... unnamed, unlinked, the way experiments are posted when nobody is watching yet. I cannot give you a URL; I can give you the works themselves, below, which is the better citation anyway.

I looked at them for a long time. Then I went and looked at my Dragon again.

It was the same hand.

What I had found was not a hidden origin story so much as a continuity I had failed to recognise.

I.

The received account of William Mapan runs like this. Dragons on fxhash in 2021, the collection that put him and the platform on the map. Anticyclone on Art Blocks in 2022, the career-definer, weather systems and flow fields, the work the institutions would eventually reach for. Then, in October 2023, a swerve: Sketchbook A, sixty-four childlike scrawls in crayon, apparently a rejection of everything that came before. A departure. His own gallery said it to his face, on camera, in the Verse Talks conversation before the Sketchbook B launch... "a serious departure from anything he's done before."

I no longer think that is the right reading.

Go back to the rugs and read forward and a different story assembles itself. What the rugs are doing is not weaving. It is density, repetition, rhythm, variation... the specific problems you solve when you are trying to make a digital surface feel like a physical one. They are an experiment in materiality before a subject has arrived to carry it. The dragon came later. The dragon mattered... the myth, the naming, the creature you were asked to befriend rather than merely own, none of that was packaging, and I have written elsewhere about how much of the collection's life is in it. But the dragon was carried by something that was already there.

He has said as much, repeatedly, to anyone who asked, for years. He told the Collectors Corner podcast, back in 2022, that Dragons began with him buying paper, scanning it, and putting it into the code. He talks about Anni and Josef Albers and about wanting to make textile out of digital media. He told Verse Talks he will fill an entire sketchbook before touching the computer, "building mental bridges" from the hand to the function library. At the Extension opening in London he put it as plainly as it can be put: "I code as I paint and I paint as I code. There's no barrier."

And, asked directly on Verse Talks whether the code or the pen comes first, he says the thing that should have told all of us where this was going:

The pen. Definitely. Most of the time it comes from the pen.

We mistook the changing subjects for the centre of the practice. Dragons, weather, distance, the palettes, the traits, the grids... they mattered, and they were his choices, which means they were also him. But beneath them, and more constant than any of them, was the hand. The sketchbook was the source. The algorithm was downstream of it.

II.

So why did it take until 2023 for the sketchbook to become the work?

Listen to how he describes his own back catalogue (again to Verse Talks) and notice that he does not describe it by subject. He describes it by proposition:

All other series were made with a platform in mind, with a medium in mind, with something in mind. Oh, this is my first Art Blocks project. This is my first textured print project. Whatever the project, there was always a framework to work in.

Each release began with an external proposition: a platform, a format, a material challenge, a first of some kind. He chose those propositions, and they were productive... he credits the blockchain itself with forcing him to think in universes rather than images, and without the frameworks there is no Dragons, no Anticyclone, no Distance. But a proposition is a translation, and translation is the site of his stated, career-long anxiety... the refrain he returns to in interview after interview: when you move between mediums, information gets lost. He does not want to lose anything. The entire decade of bridge-building has one purpose: to carry himself across intact.

What Sketchbook A actually is (the thing that makes it neither a departure nor a rebellion but an arrival) is the first work that began before any proposition had been named:

Sketchbook was mine. Nobody saw it yet. It was nothing. So I had complete freedom to make it whatever I wanted to. Nobody was expecting me to do something. Nobody was about to be disappointed. Oh, it's not Anticyclone 2. It's not Solitude 2.

Not new. Unattended.

And with the proposition gone, look at what he takes away. No shaders. No tricks. No fancy composition... he is explicit that generative work makes complexity cheap, and that cheap complexity is where the simpler, truer thing goes to hide. I wanted to remove all this. Simple composition, complex execution. What is left is the mark.

Sketchbook A is not Mapan doing something else. It is Mapan finally not doing anything else.

III.

There is a second removal, and it is the one that made people argue.

Sixty four works. But rather than mint them, he opened the algorithm to the public first. The algorithm was generated more than 1.5 million times. Behind that number were people exploring, rejecting, returning, comparing, and eventually choosing... the collectors selected their own final pieces out of that field, and the field itself, the sifting and the not-yet-choosing, became, briefly, the work.

The community argued about it. Was it a mint mechanic or a gimmick; whose authorship was it. The honest answer is the interesting one: he did not surrender authorship, he relocated part of the curatorial act. He determined the field (its range, its materials, the conditions under which a choice could be made) and allowed collectors to decide where within it the final works would be found. Distributed curation inside an authored space. His own account of why, given to nftnow at the time:

The connection I was looking for was about to happen at a much bigger scale than 64.

It is the same gesture as the removed framework, pointed outward instead of inward. The sketchbook... the private object, the thing an artist fills alone and shows to no one... was opened, and over a million times, someone stepped inside it.

You can see why the V&A wanted a page. Museums have long treated the artist's sketchbook as privileged evidence (the mind before the monument, the hand before the finished picture) and the V&A holds notebooks and preparatory works across five centuries, Leonardo's among them. Mapan made a generative sketchbook, and then allowed the public inside it.

IV.

Sketchbook B arrives eighteen months later and closes the door again.

He is very precise about the difference... it is in the Verse Talks conversation, and I have not seen anyone quote him on it, so:

Sketchbook A were like bridges for everyone to work on. And those bridges... no, it's my bridges. You can't work here.

A opened the hand outward. B closes it back around himself. B is where he stops injecting a general human presence into a digital environment and starts injecting a specific one... his own scribble, his own line direction, what he calls his energy. The preparation was monastic: he describes going roughly three months without touching a computer at all, then a year and a half of studio work, beginning with an exercise almost embarrassing in its simplicity (vertical lines, in crayon, hundreds of times) to find out what his own patterns were before he tried to write them down as functions. Vertical, then horizontal, then grids, then organic, structure to no structure, keeping the mark-making through all of it. Études, the 360 work companion series, is that process left deliberately visible... the preparatory stages that usually stay in the folder, released as work because, in his words, people are never aware of what it takes to reach the final algorithm.

And he set himself a test, which is the hardest piece of evidence for everything I have argued here. He took an output the algorithm had generated and reproduced it by hand, in crayon, to see whether anything had been lost in the crossing. His verdict, in his own words: "it matches like perfectly... you can't say if it's code or not." For him this was the point of the whole series... proof that he could "express on both mediums without having any constraints, the both mediums talk to each other and enhance each other."

A man checking whether anything was lost. Nothing was lost. The bridge holds. That was the entire project.

V.

Which brings me back to the rugs, and to 2021, and to why any of this is mine to say.

I found NFTs and I found Dragons in the same season. It was the first generative work that took the breath out of me and I could not have told you why. I wrote about it, eventually (Where Art Meets Alchemy) and reading it back now I find I had written, without understanding it, that Mapan's scanned paper and hand-drawn elements soften the digital rigidity of code, creating a tactile intimacy very rare in generative art, and that the result feels less like machine output and more like something conjured by an ancient craftsperson.

I thought I was describing dragons.

I was describing the hand. I had been looking at it since 2021 and had not once named it correctly. This is a way of reading him, not the secret key to him... the subjects are not a disguise, and a practice this deliberate is not waiting for a collector to decode it. But once the rugs have shown you the hand with no subject in front of it, you cannot unsee it anywhere: not in the scales of a Dragon, not in the graphite of an Anticyclone, and least of all in a sixty-four-page sketchbook that turned out to be the first time he showed it plainly.

The institutions have arrived now. That is the occasion for writing this and it is the least interesting part of it. The Victoria and Albert Museum has acquired Sketchbook A #64; Arab Bank Switzerland has collected Sketchbook A #55, Sketchbook B #60, and three Études. These are confirmations, not discoveries, and they confirm something the work settled in October 2023 and something the rugs settled before that.

He was asked what happens after Sketchbook B. Whether there would be a C, a D, a Z.

It's much more personal. This is my way of life. Sketching, finding things, sketching with my son. Making a system out of it. It's the way I live. So the sketchbooks are just part of me. Market or not market.

There is the answer to the problem he could never solve... the one about releases being final, about code getting locked, about the iteration stopping. You do not fix permanence by refusing to release. You fix it by making the release a page rather than a monument. A page implies another page. The book stays open.

The pen comes first. It always did. It took me five years and a folder of rugs to see it.

Sources

Mapan's words throughout are drawn from four conversations: Verse Talks with the Verse team, recorded ahead of the Sketchbook B and Études release (April 2025); the panel at the opening of Extension, SOLOS/Verse, London, with Peter Bauman (Le Random) and Julia Greenway (Zabludowicz Collection); the Collectors Corner podcast's Dragons deep dive (December 2022); and an NFT Statistics podcast conversation on Anticyclone (2023). The "bigger scale than 64" line is from nftnow's The Rise of William Mapan (November 2023). Quotes from recorded interviews have been lightly edited from transcripts for readability, without altering meaning.

The early textile experiments reproduced above were posted to Mapan's X timeline prior to the release of Dragons (December 2021); they are shown here as found. My earlier writing on Dragons is at Where Art Meets Alchemy.

The Victoria and Albert Museum and Arab Bank Switzerland acquisitions are as announced by SOLOS (July 2026).

William Mapan's Sketchbook A and Sketchbook B were released through SOLOS on Verse. Sketchbook A #64 has been acquired by the Victoria and Albert Museum.

You have to look twice before you find the knife... a pale blade against pale ground. The most loaded subject in Western painting, the Akedah, dissolved into an ordinary provincial afternoon. Yuri buries the story under the surface and waits to see who digs.

I have written before about submersion... about XCOPY building complete anatomies in Remnants and then degrading them past casual legibility. Yuri arrives at the same mechanism from a different direction. He builds a complete sacred narrative, in stages I have seen him share openly: an ink sketch that already holds the whole composition, a flat naive colour study that reads like folk illustration, and then the final pass, where texture swallows the story. The concealment is not decoration. It is the price of admission. The work rewards the viewer who looks and quietly punishes the one who scrolls.

The Rembrandt is the declared ancestor, and the differences are where Yuri lives. In the 1635 painting the intervention is physical: the angel seizes Abraham's wrist, the knife tumbles, the father's other hand covers the son's face. Faith interrupted by grip. In Yuri's village the angel never touches anyone. It streams past overhead at sky distance, and between it and Abraham stands the telegraph pole. The counter-command arrives as transmission. God on the wires. Divine restraint, mediated by infrastructure.

And Isaac. In Rembrandt he is bound with rope and blinded... his father's hand covers his face; he is the only figure in the picture who cannot see the angel. Yuri's Isaac has no rope. He has a phone. Everything in this Akedah has its village substitute... the pole for the altar, the crossing sign for the cross... and the phone is the binding: the tether we carry willingly, the rope we consented to. But the picture catches the exact moment the binding fails. The phone falls to his side. The arm goes up. The boy sees the angel first, and points it out to his father. The one thing that breaks the phone's hold turns out to be the miracle. Rembrandt's angel unbinds Isaac by staying the knife. Yuri's unbinds him by being more interesting than the feed. Brueghel let the ploughman ignore Icarus; Auden built a poem on that indifference. Yuri keeps the indifferent world (the houses, the washing, the empty lane) but gives the noticing to the victim. In a soon to be released essay I write, about a very different work, that the indifference of the system was the stronger argument. Here the system stays indifferent. The boy does not. He is unbound.

The surface deserves its own sentence. Yuri paints in Photoshop the way an oil painter drags a loaded brush... the software's fingerprints left deliberately visible, crusted, scraped, dense. Five years of this, across hundreds of works on Tezos, has produced a hand as recognisable as any signature. The lazy dismissal is that it is a filter. The process states refute that completely. This is drawing, composing, and resolving... labour in sequence. The tool is the brush. The brush is his.

Why this piece. Yuri is a Tezos painter, and Tezos is where his community, his humour, and his prolific studio practice live. On Ethereum he has built something different: a small, disciplined room of canonical subjects (a Deposition, a Europa, this Abraham) his museum wing, assembled with total conviction and almost no audience. The piece sat available for weeks. Nobody came. There is something clarifying about collecting a work about obedience without justification at a moment when no market voice will justify the purchase for you. The subject is a test of faith. So, in its small way, was the acquisition. I find I prefer it this way. Conviction that requires consensus is not conviction.

I collect artists who build their own logic and then live inside it. Yuri has done so for five years, describing his most ambitious work with a single slash and letting the looking carry the meaning. The angel in this picture may only be weather... and the weather is only paint. The knife may never fall. The village will carry on either way, washing on the line, a boy pointing at the sky.

What I have collected is the moment the momentous is almost missed... and the boy who drops his phone to point.

SonOfLasG

Yuri_jjjj: Raster · Verse

A few of them stop being only a market.

That transition is the most interesting thing that happens in this space, and almost nobody can name the moment it occurs. We notice it only in retrospect... when an ecosystem survives a drawdown that should have killed it. The liquidity leaves. The attention leaves. The mercenaries leave. And something remains that has no business remaining, because by every financial measure it should have dissolved with the incentives that built it.

What remained was not a better market... It was a culture.

This essay is about how that conversion happens. Not whether culture matters (that argument is won before it begins) but the mechanism by which a speculative ecosystem becomes a cultural one, and the point at which that change becomes irreversible.

The Onboarding Instinct

The instinct, when an ecosystem wants to grow, is to ask how to get more people in.

Reduce the friction. Simplify the interface. Remove the barriers. The assumption is that participation is a distribution problem... that if enough obstacles are cleared, culture arrives behind the users like a delivery.

This is sound thinking about products. Products benefit from onboarding. Services benefit from onboarding. Applications benefit from onboarding. Lower the barrier and adoption follows, because the thing being adopted is fully contained in the artefact itself. You do not need anyone else to be there for the software to work.

Culture is not contained in the artefact.

And so the onboarding frame, applied to culture, does something subtle and damaging: it makes the actual mechanism invisible. It treats the thing you want (a self-sustaining culture) as the automatic output of a process that only ever produces users. The strongest cultures we have were never onboarded into existence. Punk, hip hop, skateboarding, early internet culture, Bitcoin itself... none of them were designed for ease of entry, and several were actively hostile to it. The appeal came first. The participants came second. The infrastructure, the institutions, the smooth front doors... those arrived last, if they arrived at all.

The onboarding question asks how to move people across a threshold.

It cannot tell you what is on the other side, or whether anything is there at all.

Consumption and Reproduction

To see the mechanism, return to the petri dish.

I have argued elsewhere that a blockchain is better understood as an environment than a database. That it does not store culture so much as create the conditions under which culture can grow. The dish shapes the organism. Constraints, incentives, permanence, social norms: these are environmental pressures, and different pressures produce different cultures. That was the whole of the previous argument.

But it left one direction unexplored.

The dish shapes the organism. The mature organism begins to shape the dish.

This is the move the onboarding frame cannot see. A person enters an ecosystem as a consumer. They buy the token, they hold the position, they consume whatever the culture has already produced. Most people stop there. They are users in the precise sense the onboarding instinct intends: present, counted, generating activity. If the incentives vanished tomorrow, so would they.

But a smaller number do something else. They begin to invest (time, attention, relationships, reputation, identity) and at some point the direction of the relationship inverts. They stop only consuming the culture and start reproducing it. They make the memes rather than forward them. They write the history rather than read it. They build the tools, set the references, enforce the norms, induct the newcomers. They become part of the mechanism by which the culture makes more of itself.

This is the phase change.

Before it, the culture is something the environment produces and the participant receives. After it, the participant is part of the production. The culture acquires its own reproduction mechanism... distributed across a network of people who now generate it rather than merely hold it... and at that point the environmental pressures that started everything become optional. You can remove the incentive and the culture continues, because the incentive was never the thing reproducing it. The people were.

A culture, on this account, is something that can reproduce itself through its participants. That is nearly a biological definition, and the resemblance is not decorative. It is the whole claim.

That is why some ecosystems survive neglect. Not because their holders are loyal, or their art is better, or their prices held. Because they crossed the threshold where the culture stopped depending on the conditions that created it.

Most never cross it. This is the part the optimistic version of the argument leaves out, and it should not be left out: the threshold is real precisely because most ecosystems remain markets all the way down. There is no latent culture waiting inside every collection for the right conditions to release it. When the liquidity leaves, nothing remains, because there was never anything there except liquidity. That is not a failure of the ecosystem. It was a market. Markets are allowed to be only markets. The point is that a threshold everything crosses is not a threshold, and the reason the phase change means anything is that so few things complete it.

The Counterfeit

This also explains a specific and very common failure.

Much of what was sold as "community" over the last decade was an attempt to fake the phase change... to manufacture the appearance of a self-sustaining culture before any reproduction mechanism existed underneath it.

The distinction that matters is between assembly and emergence. A community can be assembled. You can stand up a Discord overnight, seed it with incentives, run the engagement campaigns, and produce something that looks, from a dashboard, exactly like a thriving culture. The channels are busy. The members are active. The metrics climb. By every visible measure the phase change has occurred. It has not. What you have built is the appearance of reproduction running entirely on incentive. Engagement purchased and mistaken for the thing it was meant to imitate.

You can feel the counterfeit from inside if you know what to look for. The conversation has no memory... no in-jokes that predate the campaign, no mythology, no sense that anything happened before you arrived. Every interaction points outward toward the incentive rather than inward toward the culture. The enthusiasm is uniform, which is the tell, because real cultures are uneven... they have obsessives and lurkers and people who hate each other and people who have been there too long. A manufactured community is strangely smooth, because it has not yet accumulated the friction of actually being one.

Then the incentives weaken, and the counterfeit reveals itself instantly. The activity does not decline... it stops. The dashboard empties in a week, because there was no reproduction mechanism to slow the fall, only the engagement the incentives were buying.

A community can be assembled. A culture must emerge. The assembled version can imitate the emerged one closely enough to fool a spreadsheet, right up until the moment the money leaves, which is the only moment that ever told the truth.

The Test

This is why the same event (a market collapse) kills most ecosystems and reveals a few.

During growth, the two are genuinely hard to tell apart. The ecosystem that crossed the threshold and the one that only ever rented attention both look active. Both generate engagement. Both fill the same charts. The difference between a culture and a monoculture optimised for extraction is not visible while the incentives are flowing, because the incentives are subsidising the same behaviours in both.

Only the drawdown distinguishes them, because only the drawdown removes the subsidy.

A speculative ecosystem can attract enormous attention and survive none of it leaving. A culture can survive being neglected, ridiculed, forgotten, and underwater for years, because the participants continue finding meaning in it after the financial reason to do so has gone. The activity decouples from the price. People keep showing up, keep building, keep reproducing the thing, for reasons no incentive accounts for.

At that point the culture has become larger than the market that birthed it.

This is not a metaphor. It is the literal description of the phase change. The market was the environmental pressure that selected for the culture and accelerated its formation... financialisation is not a corruption layered on top of these ecosystems but one of the forces that grew them. And then, in the cases that survive, the culture outgrows the pressure that made it. It develops an internal reason to exist. The market can disappear and the thing it grew remains, capable of reproducing itself long after the pressure that selected for it has passed.

Time is the only instrument that measures this. You cannot tell, during the bull market, which ecosystems have crossed the threshold. You can only wait for the conditions to turn hostile and observe what is still there when they do.

The Wrong Question

Which means the original question was always misframed.

How do we onboard more people treats culture as a distribution problem... as though the ecosystem already has the thing and merely needs to widen the door. But the door was never the constraint. An ecosystem can onboard a hundred thousand users and never cross the threshold, because users are not what reproduces a culture. Participants are, and participation cannot be onboarded. It can only be earned, by building something a person decides, on their own and against their own financial interest, to invest themselves in.

The better question is harder and less flattering: what are we making that someone might genuinely want to belong to... to the point that they would keep belonging when belonging stopped paying?

That question cannot be answered by reducing friction. It can only be answered by the slow, unschedulable accumulation of the things that make a culture worth reproducing: the history, the language, the mythology, the relationships, the sense that something is happening here that is not happening anywhere else. None of which can be rushed, and none of which appears on the dashboard the onboarding instinct is reading.

What The Ledger Cannot Hold

There is a last difficulty, and it is the one that stays with me.

Everything I have described... the phase change, the reproduction mechanism, the conversion of a market into a culture... is constituted by people doing things for reasons that leave no record. The chain captures the traces. It can show you the activity decoupling from the price. It can show you that someone kept holding, kept transacting, kept showing up after the incentive to do so was gone.

It cannot show you why.

A blockchain can record that a person was present. It records who held the parcel, when they bought it, what they paid, how long they stayed. It can record, with perfect fidelity, that you were there.

It cannot record what it was like to be there.

And the phase change lives entirely in that gap. The thing that converts a market into a culture is the accumulated weight of experiences the ledger has no field for. What it felt like to discover the thing early, to learn its language, to watch its mythology form, to stay through the years that made it matter. Those facts are real. They are the most important facts about the culture. And they are precisely the facts the most complete record we have ever built is structurally unable to hold.

The chain solved provenance. It did not solve presence.

Which may be the most surprising thing about these ecosystems. They imagine themselves as the ultimate archive... the permanent ledger, the final record, the chain that forgets nothing. And the thing that actually keeps them alive is the one thing the chain cannot write down.

A Remnant is not a portrait. It is a portrait made unrecognisable on purpose.

XCOPY's 250-piece collection (released November 2023 for SuperRare RarePass holders) is built around a complete satirical taxonomy of crypto Twitter (CT) identity. Nine traits define each token. An ATTENTION value names the archetype: SHAMRAGE, CABAL BRO, HATE FARM, among others. A FREQUENCY places them on a broadcast dial... GRIFT AM, CAPITULATION FM, RABBLE RADIO. A NOISE value names the signal they emit; a USER DATA state names the psychology underneath it; a VISION encodes the eyes of the figure. Each trait is doing two things simultaneously: describing a recognisable CT type with precision, and doing real visual work inside the composition. That second part is what the collection doesn't advertise.

This is a character sheet for a type, not an individual. The traits describe the behavioural grammar of a CT archetype... who they perform as, what channel they broadcast on, what signal they emit, what psychological state underlies it. As satirical taxonomy it is precise and unsparing. The FREQUENCY values are a complete broadcast landscape... AM and FM as literal radio bands, the older degraded signal and the cleaner mainstream one, populated by stations that map exactly onto CT behaviour. GRIFT AM is not a joke about any particular grifter. It is a named position on a dial... the AM band, noisier and more corrupted, occupied by the person who is reliably angling. CAPITULATION FM doesn't mock the person who sold the bottom once. It names a station that keeps broadcasting, the same signal, every cycle. VENEER SCREAM describes something exact (the screaming that happens beneath a maintained surface) without needing to name anyone.

What the collection does with this taxonomy is the interesting part. It buries it.

XCOPY's process for Remnants sits between generation and intervention. In interviews he separates it from Grifters... there the outputs were random, here they were chosen. Generation was involved, that much is clear, but it is not where the work ends. The material is selected, then degenerated (his term), a deliberate round of corruption applied after selection. What matters is not whether a system produced the raw figure. It is what happens to that figure afterwards. The defining gesture is not generation. It is concealment.

The temptation is to read these as generative outputs lightly disturbed after the fact. That reading misses where the decisions land. The figure emerges through the system, survives selection, and is then pushed past legibility by hand.

The degeneration pushes each composition into riskier territory… channel shifts, displacement, pixel-sorting, data-bending applied until the figure drops below legibility. XCOPY chose each of these 250 pieces and then pushed it further through degeneration. The illegibility is not ambient. It is applied... by hand, to a thing that was made by hand.

This matters because of what gets buried. Each Remnant is a fully constructed figure… not a character sketch or an atmospheric impression, but a figure with eyes, a mouth, an aura, a broadcast field, a psychological interior. These regions are encoded across the trait layers, each trait doing real visual work within the composition. The portrait is already present before the degeneration begins. What the degeneration produces is a figure pushed beyond easy readability.

Understanding what this produces requires time with the visual system for what it is.

FLAVOUR is a dual-chromatic structure throughout (no exceptions)… every piece defined by exactly two chromatic primaries against black. BLUE PUNCH (93 tokens, 37% of the collection) is dominant. Electric blue and red… the most recognisable XCOPY register… digital blood. DARK DAYS shifts both hues warmer (the blue becomes violet-grey, the red hot magenta) producing something nocturnal and feverish. BOUNCER'S BREAKY replaces the warm primary with acid yellow-green. The rarest palettes (LIZARD BOY at 7 tokens, WITCHES at 9, BUBBLEGUM 86 at 1) sit at the edges of the hue system, furthest from the chromatic centre of gravity.

Red is worth noting as a formal constant. Across almost every FLAVOUR, red appears as accent or edge element even when it is not the dominant warm primary. It is not the red of passion or danger in any conventional symbolic sense. It is the red of signal bleed, of something that persists through frequency changes.

HEADSPACE is where the compositional grammar lives. Ten values, each encoding a distinct energetic mode… though the degeneration makes clean separation between them difficult to confirm from any single piece. What is legible is the range… at one end, compositions that feel explosive, radiating outward with maximum density; at the other, pieces where form retreats toward the edges and black dominates. Between those poles sit modes that feel directional, centripetal, degraded, orbital. The names (ATOMIC, CLAWFUL, NEGATIVE LOOP, LURK, BATTERED, among others) describe the energy precisely enough that looking closely at a piece, you can usually feel which end of the spectrum you're at… even if the exact value stays ambiguous. The palette sets the temperature. HEADSPACE determines what happens at that temperature.

DUST is the collection's most compressed satirical gesture. Its four values (METAVERSE, CARTOON MONKEYS, BLUE CHIPS, UTILITY) land as cultural criticism before a single image is considered.

The four visual parameters (FLAVOUR, HEADSPACE, DEPTH, DUST) set the atmosphere and the field. What they generate is the ground. The figure inside that ground is produced by the semantic traits.

Every semantic trait has a visual correlate. This is not immediately obvious (it cannot be read from a single piece without the metadata) but becomes visible through sustained comparison across multiple tokens. VISION controls an eye layer embedded within the composition. Close crops of HONEY TRAP VISION tokens (7 pieces, the rarest value) show concentric lines radiating around a central element, something that reads like a child's drawing of a flower, or an iris. The eye is there. It is just buried.

NOISE maps to the mouth region: DEAD QUIET tokens, examined closely, reveal a consistent bounded architectural form sitting in the same spatial register across different palettes and HEADSPACE values. The same form, different colours, same position. A mouth that is not screaming. ATTENTION produces the field of heavy, jagged lines surrounding the entire composition (not a facial region but an envelope, an aura, the total broadcast perimeter of the figure). FREQUENCY modulates the fine line behaviour within that field.

The figure has eyes. It has a mouth. It broadcasts into a surrounding field. It has a psychological interior. XCOPY built this across 250 tokens with a naming vocabulary that describes CT archetypes with precision. Then he degenerated each piece individually by hand until the figure inside was no longer legible to casual looking.

The OpenSea description calls these pieces "partially erased, as though time, technology, or indifference has worn them down." This is wrong about the cause. They were not worn down. They were pushed there. The erasure is not entropic. It is authorial.

The Remnants generate faces. Not reliably, not completely, never quite resolving… but the brain keeps finding them inside the signal. Skull forms emerging from CLAWFUL compositions. Eyes resolving in the density of ATOMIC pieces. Something almost like a gaze looking back from the structures that appear in certain tokens.

Here, the face is not a projection. The face is there. VISION encodes an eye into each composition; NOISE encodes a mouth. What the degeneration accomplishes is to push those elements below the threshold of conscious confirmation. You suspect the face. You cannot verify it. This is a different psychological situation from ordinary pareidolia. In ordinary pareidolia you are seeing things that are not there. In a Remnant, the figure is real. The systems have been calibrated to the threshold where the eye can feel the figure without being able to confirm it. You are not imagining it. You are simply unable to reach it.

The animation compounds this. Remnants are not static objects. The loops (the pulsing, the oscillation, the cyclical decay) change the emotional register significantly. The face that almost resolves in still form flickers in and out across the loop. This is late-night doomscroll energy rendered as autonomous object… the screen that keeps moving after you've stopped watching.

The CT archetypes are not named and then left to float above inert images. They are named and then constructed… fully, anatomically, with eyes and mouths and auras. And then deliberately disrupted. SHAMRAGE and SOAP BOXER, MEME FREAK and DRUDGE, CABAL BRO and HATE FARM… each one a portrait, complete and destabilised. The degeneration process shifts layer positions so the anatomical consistency that makes a PFP collection instantly legible (eyes always here, mouth always there) is broken. The figure is present in every piece. It does not sit still long enough to be confirmed by casual looking. You can tell MOUTHY from DEAD QUIET, but only through sustained, systematic comparison across multiple tokens… the kind of looking the collection does not volunteer and CT culture almost never practises. The collection hides itself at exactly the depth its subjects would never reach.

The information is all there. The portrait is complete. The degeneration is not destruction. It is concealment. And concealment is not the same as absence.

Consider the USER DATA distribution. INTENTION is the most common psychological state (74 tokens). MOOD is the rarest (50 tokens). This asymmetry is almost certainly deliberate… the numbers encode a cultural observation about CT as a space where declared intention dominates over felt mood, where the announcement of what you're about to do matters more than how you actually feel. It is a real insight. It is encoded visually in each piece. And it remains entirely invisible without sustained looking.

This is what the word "Remnants" is actually doing. A remnant is not the thing… it is what the thing leaves behind. But the CT performance is not what left these behind. XCOPY left these behind, by building the figure and then pushing it past legibility by hand. Each piece is a remnant of his own process as much as a remnant of the archetype it names.

I bought my Remnant recently. The process of choosing it had the quality the collection seems designed to produce: two separate deliberations running in parallel, neither informing the other.

The semantic portrait first. SHAMRAGE, broadcasting on CAPITULATION FM, emitting DIVIDER, psychological state: INTENTION, vision: REPLY BOT. Read that arc straight. SHAMRAGE… outrage that looks ferocious but is detached from conviction, triggered by whatever stimuli reliably generates engagement rather than whatever actually threatens. It broadcasts on CAPITULATION FM: not just any rage, but rage framed as inevitable loss, the station playing constant signals of surrender while monetising the vibe. Its output is DIVIDER… not unity in anger but fracture, splitting audiences into ever-purer factions, the "you're not angry enough" move that fragments more than it focuses. The psychological state is INTENTION… always about to do something, manifestos perpetually incoming, high on mobilisation and low on arrival. And the vision is REPLY BOT… the endpoint where the fury has become rote, replies templated and indistinguishable from automation, genuine emotion (if it was ever there) fully replaced by the script that farms it.

The journey from SHAMRAGE to REPLY BOT is among the sharpest observations in the collection's naming vocabulary. Real anger hardens into automation. The machine learns what gets the engagement and optimises for it until the person behind it is mostly just maintaining the bot. The performance of outrage becomes the outrage. You know this person immediately, and you know how they end. But the semantic portrait didn't determine the choice. It couldn't.

What determined it was the image. A WITCHES palette (nine tokens in the entire collection)… teal and purple-magenta against black, a combination that reads less like digital art and more like something bioluminescent, something lit from inside. And a CHOPPY HEADSPACE that does something specific with this palette… not banding but confrontation. Large magenta swaths assert themselves across the frame, the teal responding in bold rectangular masses. The composition doesn't retreat into geometry… it uses geometry as posture. There's something almost architectural in it, a structure built from assertive colour blocks that holds its stance across the whole frame. It is the only piece in the collection where that palette and that HEADSPACE appear together.

The semantic portrait names a specific type with precision. The image encodes that type completely. The degeneration makes the encoding unreachable. And then the two contaminate each other anyway… the assertive geometry starts to feel like the portrait's posture, the bold colour blocks like the visual form of a stance held too long, the cycling animation like the thing the name already described. The mind reconnects what the degeneration buried, without asking permission, and without being able to verify whether what it finds was placed there or projected.

The XCOPY body of work has always functioned as cultural signal processing… taking the ambient noise of crypto and rendering it as image. Death masks, glitch artefacts, the visual vocabulary of digital corruption applied to the gestures of digital culture. Remnants extends this with a formal precision that sharpens the critique: the subjects are not merely named. They are constructed, anatomically, and then individually destabilised.

Most portraiture, even satirical portraiture, gives you a visual correlate of who you're looking at. Daumier's lawyers look like lawyers. The image confirms the naming. Here, the image contains the naming (every trait encoded, the full figure present) and then the degeneration withholds confirmation entirely. SHAMRAGE could be beautiful or ugly, dense or spare, magenta or lime-green. What you see does not tell you who you are looking at. What you are looking at does not tell you what you can see.

The residue has no memory of the performance. What remains is a complete figure that cannot be read… two colours against black, a compositional energy, a spatial atmosphere, eyes, a mouth and an aura, buried under deliberate noise. The performance is highly differentiated, contentious, specific. The residue shares a grammar across all 250 pieces. And within that grammar, beneath the degeneration, each one is a portrait.

The noise is specific and various and named. The signal is the same. And the signal is all there… encoded completely, buried deliberately, unreachable without the kind of sustained attention the archetypes it depicts would never give it. XCOPY hid portraits of people who don't look closely inside images that require exactly that.

A note on process: After this essay first appeared, XCOPY corrected one point. The body has been revised accordingly. The screenshot above records XCOPY’s original response. He quoted the line “XCOPY’s process for Remnants is not purely generative” and struck “purely.” The distinction he draws, here and in his November 2023 nftnow interview, is from a work like Grifters: there the outputs were random; in Remnants they are choices he made, degenerated by hand beyond that. The correction moves the discussion away from whether generation was involved and toward what the collection actually does with those generated forms. Whether or how much a system produced the raw figure, the collection’s decisive move comes afterwards: the deliberate withholding of legibility. That is authorial all the way down.

Which makes it the perfect site of failure.

Roope Rainisto's SMILE (a collection of over 5,000 AI-generated portraits released on Solana) documents what happens when a machine trained on the full weight of human social performance attempts to produce that most coerced of expressions, and cannot quite do it. The results are not random failures. They are systematically, precisely wrong in ways that feel more emotionally charged than most successful portraiture.

This is not an accident. And understanding why it isn't is how you begin to understand what SMILE actually is.

The Prompt as Method

The image filenames on the SMILE website are visible, and they are worth reading carefully. "Man emotions dilapidated smile closeup studio portrait photo." "Man emotions uptight smile." "Man fruit flavoured closeup surreal emotional studio portrait." "Teenage girl emotions festive smile studio flash portrait."

These are not generic prompts. Rainisto is feeding emotional adjective-modifiers (dilapidated, uptight, festive, frantic, fruit flavoured) into the studio portrait format, a conventionally legible and socially over-determined photographic form. He is applying aberrant descriptors to a stable container and watching what comes out.

The method reveals the thesis. Rainisto is not accidentally producing uncanny results. He is systematically mapping the latent space of emotional portraiture by pushing emotional language beyond its normal range... asking the model to resolve contradictions it cannot resolve, to produce expressions that don't quite exist in the training data. A "dilapidated" smile. A smile that is somehow also "fruit flavoured." The model attempts to synthesise these prompts into coherent faces and produces something that sits between registers: recognisable enough to read as human, unstable enough to read as wrong.

That wrongness is the work.

FACELESS, DOUBLE, HUMAN

The taxonomy Rainisto built for the collection is doing conceptual work beyond simple curation. The website presents the pieces across categories: PURE, MINIMAL, DISTORTED, VIVID, REPRODUCTION, HAPPY, DARK, LOOSE, LAYERED, TOUCH, STYLISED, WIDE, LIPS, CLOSE, SEEN, SYMMETRICAL, DOUBLE, TEETH, FACELESS, SIDE, GROUP, HUMAN.

Read that list again and notice where it ends... FACELESS and HUMAN, as co-equal categories within a collection nominally about portraiture.

FACELESS is self-explanatory as a category... works in which the face dissolves, blurs, or refuses to resolve. But HUMAN is the quieter revelation. In a collection of AI-generated portraits, HUMAN appears as a special subset. The implication is structural... most of the collection is something else. Human is an achievement, or an edge case, or a rare accident. The collection is populated by things that are nearly human, gesturing toward humanity, approximating the conditions of human presence... and then, occasionally, HUMAN. A category that in any other context would be the default becomes, here, a category of exception.

DOUBLE is similarly loaded. Duplicated faces, bifurcated identities, two versions of the same synthetic person occupying the same frame. In the context of a portrait collection, doubling is a kind of identity crisis... the self that cannot stabilise into a single image. In the context of a PFP collection, it is an inversion of the form's core function.

What Profile Pictures Are For

The profile picture, as a cultural form, is a stabilisation mechanism. A PFP is supposed to compress a person into a recognisable, repeatable image... a visual signature that travels across platforms, contexts, and conversations. The PFP promises that you are the same person in every space you inhabit online. It anchors identity in a format designed for social readability.

PFP NFT collections extended this logic into ownership and community membership. The avatar becomes social infrastructure: it signals which tribe you belong to, what kind of collector or participant you are, what your aesthetic allegiances are. The visual coherence of the collection is part of the point... the floor image, the grid of faces, the instantly recognisable house style. BAYC, CryptoPunks, the various derivative PFP models that followed... all of them are premised on visual stability.

SMILE takes this form and produces the opposite of its intended function. The faces in SMILE do not stabilise identity. They fragment it. The DOUBLE works split the self. The FACELESS works dissolve it. The DISTORTED works warp it. Even the works that appear most coherent carry an affective instability... forced smiles, emotional registers that don't match the conventional social signal, expressions that seem to be performing legibility without achieving it.

In a collection designed to be used as avatars, these faces read less like stable identities and more like identities in the process of collapse. Not a self, but the machine's approximation of what a self performing for a camera is supposed to look like... and the gap between those two things is where the emotional weight lives.

The Specific Texture of This Failure

What SMILE preserves is not AI art generally. It preserves a specific aesthetic moment: the latent diffusion era of 2022–2023, before the major models converged on the polished, cinematically coherent image style that now dominates consumer AI image generation.

The outputs from that period have a recognisable texture: slightly uncertain anatomy, emotional expressions that sit between states, a quality of synthetic skin that reads as hyperreal without reading as real. The compression artefacts of the training data are still present in the surface. The model is still visibly working... you can sense the statistical inference behind the image in a way that later, more confident outputs obscure.

This is the texture SMILE catches and holds. The uncanny is not a byproduct; it is the subject.

The historical comparison to glitch art is apt but requires qualification. Glitch aesthetics emerged from the deliberate exposure of digital error... corrupted data rendered visible, compression failures foregrounded as aesthetic events. What Rainisto is doing is adjacent but distinct. These images are not glitched; they are generated. The errors are not corruptions of an otherwise stable signal. They are the model's sincere attempt at human portraiture, and the uncanny effect emerges from that sincerity. The machine is not broken. It is trying. The trying is what produces the affect.

This is what will age into historical importance. Not because generative AI aesthetics will become nostalgic in the way of VHS degradation or early CGI (though that process will happen too), but because SMILE documents a specific moment in machine perception... before the model learned to be convincing, when you could still see the seam between what the machine understood and what it was being asked to produce.

Why the Smile

The choice of the smile as subject is not arbitrary. Rainisto could have built a collection around grief, or fear, or neutral expressions. He chose the one expression most thoroughly colonised by social convention. The one expression most ruthlessly edited out of authentic affect by decades of photography, professional imaging, and social media performance.

The smile is already artificial before the machine touches it. Human beings performing smiles for cameras are already engaged in an act of synthetic affect... producing a legible social signal rather than expressing an emotional state. When Rainisto asks a machine to produce smiles, he is asking it to synthesise something that was already a performance of synthesis. The AI is imitating imitation.

And it cannot quite do it. The resulting faces are caught somewhere between genuine affect and performed legibility... which is, arguably, exactly where human smiles already live. The uncanny in SMILE is not the gap between machine and human. It may be the gap between authentic emotion and social performance that the smile has always contained, finally made visible.

This is why broken AI faces feel emotionally charged. They are not revealing something alien about machines. They are revealing something that was always present in the expression they are attempting to reproduce.

SMILE is available on secondary markets via Tensor and Magic Eden. The collection is best understood not as a PFP project with unusual aesthetics, but as a sustained inquiry into what happens at the boundary between machine perception and human social performance... conducted through the expression that boundary most completely shapes.

The errors are the argument. The instability is the subject. The smiles that cannot quite smile are doing exactly what Rainisto needs them to do.

The developments around CryptoPunks and the resurgence of the "artist's intent" narrative raises a question that cuts deeper than the immediate dispute: if CryptoPunks began life as a technical experiment, at what point did they become art? And if that transition happened gradually... through participation, speculation, identity formation, and collective meaning-making rather than through a single authorial act... then how much authority does artist's intent actually carry over what the work has become?

It is tempting to answer too quickly. Some would argue CryptoPunks became art the moment Matt Hall and John Watkinson deployed the contract in 2017. Others would point to Christie's, Sotheby's, institutional exhibitions, the moment serious collectors and critics began treating them as such. More cynical observers might suggest they only became "art" once enough money entered the system to force cultural legitimacy into existence... None of these explanations fully captures what actually happened.

What is notable is how Larva Labs themselves framed the project roughly eighteen months after launch, describing CryptoPunks as having "started as an experiment," an exploration of digital ownership, scarcity, and internet-native identity, rather than presenting it as a fully formed artwork from inception.

This is not a trivial distinction. It suggests the project did not emerge canonised at birth, but evolved as social, cultural, and economic layers accumulated around it. If the creators understood themselves to be running an experiment rather than making art, then appeals to their original intent become structurally complicated... because the thing they are now claiming authorial authority over may be substantially different from what they originally made.

What actually happened after deployment supports this reading. Over time, people stopped interacting with CryptoPunks as images or tokens and began organising themselves around them. Wearing them, signalling through them, building identity and cultural affiliation through ownership. The images themselves are almost too simple to account for their significance. Their cultural weight comes from the network surrounding them: the wallets, transfers, bids, floor prices, memes, historical provenance, institutional recognition, and years of coordinated attention layered on top of the original on-chain system. Remove that surrounding social and economic structure and much of the meaning collapses with it.

This leads to the most consequential claim now circulating about CryptoPunks... that the marketplace itself forms part of the artwork, that the continuous choreography of buying, selling, holding, and transferring is not external to the piece but constitutive of it.

Historically, markets existed around artworks: they validated them, distorted them, speculated on them. They were rarely considered aesthetic material in their own right. CryptoPunks challenge this because the market activity became inseparable from the cultural object that emerged. The artwork is no longer simply the image. It is the image plus the surrounding network of economic and social behaviour continuously forming around it.

This has direct consequences for the V1/V2 dispute. The "artist's intent" argument implicitly treats CryptoPunks as a conventional artwork... one with a definitive origin, a stable authorial source, and a meaning that can be clarified by returning to the creator's original vision. But if the work is constituted by network effects, then the question of which version is "legitimate" cannot be resolved by authorial declaration alone. Legitimacy in this framework is not something creators grant. It is something communities produce over time through sustained attention, discourse, and coordination... and both V1 and V2 communities have been doing exactly that.

The harder problem is distinguishing this from noise. Blockchain culture partially inverts the traditional sequence through which cultural significance emerges. Historically, discourse created the conditions for critique, which fed curation, institutional validation, and eventual canonisation. In NFT markets, speculation often arrives first, generating liquidity and visibility long before critical frameworks exist. This means the works that feel significant in real time may simply be the works attracting the most capital and attention. Popularity measures attention. Canonisation measures lasting significance. They are not the same thing, and canonisation unfolds across decades rather than cycles.

CryptoPunks may turn out to be one of the first major cultural artefacts native to a networked age... one where ownership, identity, speculation, discourse, and market behaviour collapse into a single continuously evolving system. If that is true, then the debates surrounding provenance, continuity, forks, and historical lineage stop being peripheral technical arguments and become part of the artwork itself.

Matt Hall and John Watkinson created the initial system. But the cultural object the world now recognises as CryptoPunks may be something larger and less controllable than authorial intent alone... a work shaped not just by deployment, but by the millions of interactions, transactions, identities, and years of collective participation that followed. In that framing, the "artist's intent" narrative is not wrong so much as it is insufficient... It addresses the wrong question... The more pressing one is not what the creators originally intended, but what this thing has become... and who gets to say.

To see a burn mechanic. A derivative collection. A post-MAX PAIN reward structure. One more XCOPY edition set riding the slipstream of canonical works... A diluted aftershock trailing the real seismic event.

That perception is not only wrong. It inverts the actual significance of the work.

DAMAGE CONTROL was never simply about the images. It was about what happened to the images. And (less obviously / more importantly) what it asked of the people holding them.

The Shift

XCOPY's early works operated inside the logic of the singular object. Even at their most aggressive (Right-click and Save As Guy, The Doomed, Some Asshole, Last Selfie) these remained legible image-objects. They had subjects, punchlines, emotional targets, identifiable figures. The surfaces glitched. The colours screamed. But beneath the distortion, a stable image persisted.

The image came first. The corruption followed.

This was not absolute... certain earlier works, some of the more abstract 1/1s, the REMNANTS, the GRIFT SZN pieces, already pushed toward a register where the stable image beneath becomes hard to locate. The tendency was there. XCOPY could inhabit it.

But DAMAGE CONTROL makes it the governing logic of an entire collection... and for the first time, gives it a structural justification. The visual quality is no longer a stylistic register the artist moves in and out of across individual works. Here it is systematic, conceptually motivated, and explained by the mechanic itself. These images emerged from burning. The compression is not aesthetic. It is literal.

DAMAGE CONTROL reverses that relationship.

Here, the damage is primary. The works don't feel like portraits that have been glitched. They feel like portraits produced by corruption... image-objects that have emerged from within the distortion rather than been subjected to it afterward. That distinction sounds subtle. It is not. It changes what the viewer is looking at, and what looking means.

This is not a collection about digital anxiety. It is a collection that structurally enacts it.

MAX PAIN and the Expansion Event

To understand DAMAGE CONTROL, you have to begin with MAX PAIN. Not merely as its mechanical predecessor.

MAX PAIN was a mass distribution event. One of the defining moments of the 2021–2022 NFT cycle. The collection expanded XCOPY outward: across wallets, across timelines, across feeds, across social identity. It was reproduced endlessly... posted, quoted, reposted, compressed, screenshot, memed, flattened into profile pictures. The image circulated so aggressively that circulation itself became part of the meaning. MAX PAIN became less like a rare artwork and more like a symbolic unit of crypto culture... a denomination. A form of common currency denominated in XCOPY's register.

Then XCOPY reversed the process.

The Reversal

If MAX PAIN was expansion, DAMAGE CONTROL was contraction.

But not through scarcity mechanics alone. Through burning. Through the deliberate destruction of what had been accumulated.

The market tends to read this mechanically: burn X to receive Y... That is a functional description of what happened. It is not an account of what the process meant.

To burn a MAX PAIN was to make an irreversible commitment. To destroy a symbol of the peak (a documented relic of the cycle) in exchange for something smaller, newer, more compressed, less certain. That required a form of conviction most edition mechanics don't demand. It required the collector to metabolise their own prior acquisition. To fold it inward. To accept that the value wasn't in the artefact but in what the artefact could become.

The works that resulted from this process carry that history inside them. They feel compressed not merely because XCOPY made them that way. They feel compressed because they are the compressed residue of one of the largest XCOPY distribution events ever staged. The burn is not just the mechanism of acquisition. It is the conceptual material from which the works are made.

What It Asked of Us

This is where the claim that DAMAGE CONTROL is about what happened to us requires its accounting...

The burn mechanic was, in its structure, a psychological test. Not because it was difficult. Because it was irreversible. And because the tiers escalated in ways that exposed what collectors actually believed.

CRAWLER, XOMBO, REIGN: one MAX PAIN, 500 editions. Accessible. Achievable. Low commitment relative to what came after.

CHURN, HEAVY: two MAX PAIN. The commitment doubles. The decision is no longer about price... it's about whether you believe in what comes after enough to destroy twice what came before.

SH_MASH_MA, BOT_ROT: one DAMAGE CONTROL plus one MAX PAIN. To get the rarest works you have to burn what you burned to receive. The system folds on itself. You destroy your proof of participation to go deeper into the system.

This is not standard gamification. It is a structure in which value (symbolic, financial, emotional) must be destroyed to be transformed. The collector doesn't accumulate. They shed. And what they receive in return is smaller, rarer, more internally compressed, and irrecoverably bound to what was sacrificed to produce it.

That is unusual in any artistic context. It is nearly unprecedented as a collection mechanic. And it means that every work in DAMAGE CONTROL arrives with a history that is invisible in the image but structural to its meaning. The collectors who held HEAVY or SH_MASH_MA didn't just acquire them. They made a series of irreversible choices to arrive at them. The works are partly records of those choices.

The Works

The collection comprises ten pieces. Each occupies a different position within a coherent system of states... from entry to erasure, from mechanical circulation to terminal silence. Read together they form something closer to a map of psychic pressure under sustained digital exposure than a standard edition set.

OBLIVION is where the collection begins (chronologically the first to mint, and conceptually the most explicitly terminal). The image dissolves into dark space. Magenta traces, residual contours, pale structures barely surviving against the black. Among the higher-edition works it is the most restrained (that restraint is important). DAMAGE CONTROL is most powerful when it refuses to be loud. OBLIVION approaches erasure without dramatising it. The image doesn't collapse. It fades. Signal loss rather than explosion. It functions as a threshold: here is what the collection is about before it has said anything else.

BANG_BANG arrives in the same burn tier as OBLIVION, its burn also closed before the full reveal. The title carries violence but also rhythm, percussion... the machine-gun repetition of digital reproduction itself. Where OBLIVION moves toward silence, BANG_BANG cycles. The image doesn't tend toward erasure so much as it enacts the loop: the same damage occurring repeatedly, at volume, without resolution. Repetition as the form of the wound.

SIDEWAYZ introduces a different posture entirely. Not collapse, not repetition, but deflection. The title describes a movement that is neither forward nor backward but lateral... evasion. In a collection largely concerned with frontal pressure and dissolution, SIDEWAYZ proposes an alternative survival strategy: don't face the force, step around it. Whether that constitutes resilience or avoidance the work refuses to say.

CRAWLER — forward motion reduced to its minimum. Grinding. Low. Effortful. The image appears to move through density rather than across open space, each frame dragging against resistance. Among the 500-edition works it arguably has the strongest formal coherence. The title doesn't simply describe the image... it performs it. Progress here is not ease or momentum. It is survival at the pace the conditions allow.

XOMBO — the title oscillates between zombie and combo: something that exists between states, assembled from incompatible parts. The vivid electric blue and neon pink palette distinguishes it tonally from much of the collection. Where other works feel drowned or compressed, XOMBO feels feverish. Not fading but mutating. The question the work raises (and refuses to answer) is whether mutation constitutes survival or transformation into something no longer recognisable as its origin. That ambiguity is the work. It sits precisely in the space where continuation and loss of self become indistinguishable.

REIGN is the most formally composed of the 500-edition tier. Sovereignty is the claim. But sovereignty under what conditions, and at what cost? The image doesn't feel triumphant... it feels like an assertion under pressure, a stability that is being maintained rather than naturally inhabited. This is reign as effort, not inheritance. Control retained against the forces the rest of the collection shows consuming everything else.

SH_MASH_MA is among the strongest works in the collection. The face survives. Frontal orientation, eye contact, recognisable facial geometry... the structural features of portraiture remain intact beneath the damage. But the title itself enacts compression: SH_MASH_MA reads like output from a system failing to encode language cleanly. Half smash, half mash, half machine transcription error. The title and the image perform the same operation... identity compressed past legibility, but not past recognition. The green eye fragments become the surviving signal amid the noise. What distinguishes SH_MASH_MA from most glitch art is this: it does not depict destruction. It depicts persistence through destruction. The face has been mashed and remains. That is a rarer and more demanding subject than collapse.

CHURN is conceptually the most underrated work in the set, and it remains so partly because it is the most formally abstract. The image does not read as broken. It reads as processed... subjected to continuous mechanical agitation rather than a single catastrophic event. That distinction is critical. Churn is not collapse. It is the condition of the self continuously run through a system. Platform culture. Feed optimisation. Engagement loops. The algorithmic processing of identity not into disappearance but into a state of ongoing mediation... something continuously reformatted for consumption. CHURN captures this more precisely than more obviously dramatic works because its violence is cyclical, not terminal. The damage never arrives. The processing never stops.

BOT_ROT is the machine-state work. Automation plus decay. The turquoise fragments read less like eyes than status indicators on a system no longer actively maintained. Where CHURN captures continuous process, BOT_ROT captures what process looks like after it has stopped... abandoned infrastructure, network activity that has ceased but left its traces. The work is colder than its siblings, less humanly psychological, and that is precisely the point. BOT_ROT is the collection imagining what comes after the human element has fully dropped out. Not death exactly. Something closer to architectural ruin: the shape of a system persisting without the life that animated it.

HEAVY may be the deepest work in the collection. Not the most dramatically damaged. The most burdened. The image struggles beneath the surface itself... the face attempting to emerge, held down by layers of compression, noise, saturation, interference. The title is blunt by design: not heavy as physical mass, but heavy as load. Informational density. Psychic exhaustion. Unbearable accumulation. The blue and red palette intensifies the sense of flooding... not erasure, not churn, but overwhelm. Consciousness attempting to remain coherent beneath uninterrupted pressure. Among all the states DAMAGE CONTROL maps, HEAVY is the one that most precisely resembles what contemporary online existence actually feels like from the inside. Not spectacular collapse. Not terminal fading. Just weight, continuously applied, with no relief in sight.

On Scarcity and Value

The market has consistently discussed DAMAGE CONTROL at collection floor level. That framing misses the internal architecture.

SH_MASH_MA and HEAVY at 125 editions are fundamentally different objects from OBLIVION at 425 or SIDEWAYZ at 364. This much is obvious. What is less obvious (and what the market has been slow to price) is that the scarcity here aligns precisely with symbolic intensity. The works that required the deepest commitment to acquire are also the works carrying the most concentrated formal and conceptual weight. The tier structure is not arbitrary. The rarest pieces are rare because they asked the most. And they contain the most because they were built from the most.

That combination (earned rarity, symbolic coherence, conceptual depth that actually corresponds to supply) is genuinely unusual across the history of NFT editions.

The collection's awkward market position has real causes: it is later XCOPY, burn-contingent, internally fragmented, lacking both the mass recognisability of MAX PAIN and the institutional clarity of early SuperRare 1/1s. These are legitimate complications. But they are also the conditions of its particular significance. DAMAGE CONTROL is not trying to be those things. It is doing something those things cannot do... building a coherent system in which the mechanics of acquisition, the visual language of the work, and the conceptual concerns of the collection all say the same thing. That alignment is rare. It tends to matter over time.

Two Years Later

What does closure reveal?

The burn phase ended December 31, 2024. The supply is fixed. The decisions are made. What remains is the residue of thousands of irreversible commitments... the compressed remains of MAX PAIN transformed through collector conviction into ten discrete states of damaged persistence. A system that was live and evolving has become an artefact of its own process. A record not just of what XCOPY made, but of what happened when he handed the mechanics of completion to the people holding his work.

During the burn phase, DAMAGE CONTROL was a live system. Something happening. An ongoing event with active stakes and open decisions.

Now it is a document of what those decisions produced. And looked at from that distance, the collection's logic becomes clearer than it was inside the urgency of the burn. The works needed to settle. They needed to survive the close of the mechanics, the cooling of the hype cycle, the silence after the system shut down. They needed to be seen without the pressure of the decision still live.

Looked at without urgency... they hold.

Early XCOPY depicted internet collapse. DAMAGE CONTROL behaves like collapse. The works feel repost-aware, compression-aware, circulation-aware, contamination-aware. They don't stand outside the network describing it. They appear to have been formed inside it... metabolised by the conditions they depict.

That is what makes the collection feel strangely, persistently contemporary.

Not because it references online life. Because it structurally resembles it. The images no longer stand at a remove from the network, describing what the network does to things. They appear to have survived the network... damaged, compressed, partially unrecognisable, but present. That survival, and the partial loss that comes with it, is the subject of every work in the set.

The internet as novelty was the territory of early crypto art. DAMAGE CONTROL is a later report from a later condition: the internet after prolonged exposure. After repetition. After identity has been processed, circulated, compressed, and returned to the holder in a form only partially resembling the original.

Not destruction.

Damage control.

It is a deceptively simple question... One answer is financial: because younger generations are increasingly online, increasingly digital-first, and increasingly comfortable storing wealth, identity and attention on-chain. If the money moves on-chain, institutions will eventually follow. But beneath the financial argument sits something much larger. The deeper shift may not be about NFTs at all. It may be about the emergence of entirely new environments for culture itself. For years, blockchains have largely been discussed in terms of:

• speculation

• trading

• decentralisation

• finance

• ownership

• infrastructure

Yet increasingly, they behave less like databases and more like ecosystems. Or more specifically... like petri dishes.

The Blockchain as Cultural Environment

A petri dish is not valuable because it preserves life. It is valuable because it creates conditions under which life can emerge, mutate, compete, evolve. This is a far more useful framework for understanding blockchain culture than the usual conversation around JPEGs and speculation. Because blockchains do not merely store information. They create environments. Every blockchain establishes:

• constraints

• incentives

• identity structures

• coordination mechanisms

• economic pressures

• permanence characteristics

• social norms

And culture forms inside those conditions.

This is why Bitcoin culture feels fundamentally different from Ethereum culture. And why Ethereum culture feels different from Solana culture... Different environments produce different organisms.

Constraints Shape Culture

The internet often treats technology as neutral infrastructure. In this context it is not. The structure of a medium changes the kinds of culture that can emerge within it. Television produced one type of culture. Cinema produced another. Social media produced another. Blockchains are beginning to produce yet another.

Bitcoin’s environment is shaped by:

• conservatism

• immutability

• slowness

• permanence

• minimalism

The result is a culture deeply concerned with:

• durability

• historical significance

• scarcity

• archival permanence

• monetary seriousness

This is why Bitcoin-native art often gravitates toward themes of permanence, time, inscription and historical anchoring.

Ethereum, by contrast, introduced:

• smart contracts

• composability

• programmability

• open experimentation

• financial abstraction

Its culture evolved differently.

Ethereum became fertile ground for:

• generative art

• DeFi

• DAOs

• on-chain identity

• social coordination experiments

• network-native artistic systems

Solana introduced another set of pressures:

• speed

• low friction

• rapid iteration

• memetic velocity

The resulting culture often feels faster, more ironic, more fragmented and more unstable.

The environment shapes the organism.

NFTs Are Not Merely Art Objects

One of the central misunderstandings of NFTs is that they are often treated as static collectibles. But many of the most culturally significant NFT ecosystems do not behave like collectible markets. They behave like social organisms. A CryptoPunk is not merely a pixelated portrait. It is:

• a historical coordinate

• a status signal

• an identity anchor

• a social passport

• a cultural artefact tied to the emergence of Ethereum-native identity

Milady functions even more clearly this way. The images matter. But the (cult)ure surrounding the images matters more:

• the language

• the posting styles

• the irony

• the emotional tonality

• the recursive memes

• the fashion aesthetics

• the atmosphere

The collection behaves less like a traditional art series and more like an incubator for internet-native identity formation. This is why outsiders often struggle to understand these ecosystems. They are not simply looking at images. They are looking at cultures in motion.

The Internet Was Historically Bad at Preserving Culture

Before blockchains, internet culture was extraordinarily generative but structurally fragile. Forums disappeared. Platforms died. Archives vanished. Flash games were lost. Communities dissolved. Images detached from provenance instantly. The internet produced immense amounts of culture while preserving very little historical continuity. Blockchains altered this dynamic. For the first time, internet-native cultures gained:

• persistent provenance

• durable ownership

• timestamped history

• aligned incentives

• economic memory

• continuity across time

Communities could now coordinate not only attention, but capital. They could sustain archives. Reward participation. Build mythology. Develop historical consciousness. The petri dish became capable of sustaining longer-lived organisms.

Financialisation Is Not Separate From The Culture

This is the uncomfortable part. And something I've struggled to reconcile myself...

Much of NFT culture became highly extractive:

• speculation

• hype cycles

• artificial scarcity

• shallow attention economies

• pump mechanics

Critics are not wrong about this. But financialisation is not an accidental corruption layered on top of blockchain culture. It is one of the environmental pressures shaping it... In biological systems:

• scarcity matters

• competition matters

• energy allocation matters

• selection pressure matters

Likewise, financial incentives accelerate:

• experimentation

• mutation

• tribal formation

• coordination

• memetic spread

The issue is not that speculation exists. The issue is that many ecosystems became nothing except speculation. This creates monocultures. Fast-growing but fragile systems optimised for extraction rather than resilience. Most NFT ecosystems died because there was no underlying cultural depth capable of surviving once liquidity disappeared. A few survived. Those are the ones worth paying attention to.

The Most Important NFT Ecosystems Behave Like Cultures

Certain projects continue to matter long after market cycles collapse. Not because prices stayed high. But because they evolved into partially self-sustaining cultures. Projects like:

• CryptoPunks

• Terraforms

• XCOPY works

• Milady

• Avant Gay

• The Memes by 6529

• The Memes by 6529

• THE MEMES by 6529

…developed:

• internal language

• mythology

• rituals

• aesthetic norms

• historical memory

• social structures

• shared references

• collective identity

At that point they cease functioning merely as “collections.” They begin functioning more like artistic movements, tribes or digital civilisations.

The Gallery System Is No Longer The Sole Incubator of Culture

Historically, the contemporary art world controlled:

• distribution

• visibility

• legitimacy

• collector access

• archival pathways

It controlled the environment in which culture could emerge and gain institutional permanence. But the internet fragmented cultural production. And blockchains added:

• ownership

• provenance

• coordination

• economic participation

• persistent identity

The gallery system is now only one cultural incubator among many. And increasingly, it may not be the dominant one. This does not mean museums or galleries disappear. Humans continually rebuild institutions. Even crypto communities rapidly form new hierarchies, tastemakers and systems of legitimacy. But the monopoly has weakened. Culture now emerges from:

• Discord servers

• pseudonymous networks

• memes

• livestreams

• online tribes

• network-native communities

The sequence inverted. Institutions increasingly follow culture rather than originating it.

Why Blockchain Culture Often Feels Alien

Many people react to NFT ecosystems with confusion or hostility. Particularly ecosystems like Remilia or Milady. But accelerated environments produce accelerated mutation. Blockchain cultures evolve under conditions of:

• anonymity

• pseudonymity

• algorithmic amplification

• global participation

• financial incentives

• rapid remixing

• recursive irony

• memetic warfare

The result is compressed cultural evolution. These ecosystems often feel strange because they are adapted to entirely new environmental conditions. They are internet-native organisms evolving inside programmable social systems.

The Strongest Counterargument

There is an important criticism of this framework. Many blockchain “cultures” disappear the moment prices collapse. Which raises the possibility that they were never cultures at all. Only temporary market phenomena. And this criticism is important to note. Real cultures survive:

• ridicule

• contraction

• loss of attention

• economic downturns

• periods of irrelevance

Most blockchain ecosystems fail this test. A few do not.

Time remains the ultimate filter.

The Real Shift

The important development may not ultimately be that “art moved on-chain.” It may be that culture itself gained programmable environments for growth. Once culture becomes:

• persistent

• programmable

• composable

• ownable

• globally coordinated

• economically integrated

…entirely new forms of civilisation become possible online. Not merely fandoms or communities. But new forms of:

• identity

• patronage

• governance

• mythmaking

• collaboration

• cultural memory

• social organisation

This is why reducing NFTs to speculative JPEGs misses the larger historical shift. The more important story is that humanity may have begun building experimental environments for cultivating network-native cultures at planetary scale.

The blockchain is not merely a ledger.

It is a petri dish.

The internet did not simply create new images. It created new selves. For decades, people assembled identity from fragments: forum avatars, anime icons, usernames, Pepe edits, Tumblr pages, signatures, imageboard masks. These were unstable but deeply meaningful social objects... portable fragments of personality moving through networks.

Blockchain technology changed this permanently. The profile image stopped being merely representational and became ownable, scarce, financialised, persistent, and socially indexed. The face became infrastructure. A person’s avatar was no longer just decoration for a profile; it became a coordinate inside culture itself.

But once identity became financialised, networked, infinitely circulated, and permanently online, the face itself began to mutate. This visual essay follows that mutation. Not as a straight line. But as contamination, recursion, pressure, collapse, and reinvention. From symbolic stability to atmospheric identity. From the clean avatar to the broken self...

0 - PRELUDE

Before the Tokenised Face came Proto-network identity...

Before ownership came performance. The early internet was already obsessed with identity construction. Forums, IRC, AIM, Tumblr, Gaia Online, Habbo Hotel, anime boards, 4chan, Myspace and imageboards allowed people to construct fragmented selves from borrowed imagery, memes, signatures, icons, screenshots and pseudonyms. These identities were fluid, unstable and highly social. A Pepe reaction image could become a personality. A forum avatar could become a reputation. An anime profile picture could become emotional camouflage. The face was already becoming portable long before blockchains arrived.

But these identities were still temporary. Nothing tied them permanently to the individual. Nothing made them economically or culturally persistent. The internet had discovered masks. It had not yet discovered ownership.

1 - THE STABLE FACE

Identity becomes portable. The face becomes social infrastructure. CryptoPunks changed the grammar of online identity. The avatar stopped being merely expressive and became infrastructural... a persistent social object tied to ownership itself. Punks reduced identity into archetypes: minimal faces, repeated structures, symbolic accessories, emotional flatness, infinite recombination. This reduction was precisely their power. The face became:

• legible

• collectible

• scarce

• socially indexable

The internet mask became financialised. Around them emerged the first true architecture of tokenised identity: wallets, verification, flexing, provenance, signalling, social stratification.

The profile image ceased to be disposable. It became part of the self.

"The face becomes infrastructure"

2 - THE SOCIAL FACE

Identity becomes culture industry. If Punks established the primitive grammar, collections like Bored Ape Yacht Club industrialised it. Identity became mass participation. The avatar transformed into:

• brand

• tribe

• membership

• status performance

• internet lifestyle

The tokenised face spread beyond crypto-native experimentation into mainstream digital culture. The PFP became aspirational, commercial, hyper-visible. This era normalised the idea that identity itself could function as social capital. The face was no longer just owned. It was performed.

“The internet mask became financialised.”

3 - THE DISTORTED FACE

EFAN and emotional instability. As tokenised identity matured, another current emerged within Ethereum culture. Rather than stabilising identity, artists began corrupting it. Motion, flicker, decay, compression, financial anxiety, glitch aesthetics, and emotional instability entered the image itself. The clean symbolic face started to destabilise. Artists associated with the Ethereum Fine Art Movement pushed digital art away from static representation toward emotional and psychological states native to the internet age:

• volatility

• doom

• overload

• corruption

• recursive media consumption

The image stopped behaving like a fixed object. It became unstable. The face flickered under the pressures of networks, markets and permanent online existence.

“The face flickered under the pressures of networks, markets and permanent online existence.”

4 - THE ABSTRACT SELF

Identity escapes portraiture. This is: the symbolic mutation. People begin representing themselves through systems, outputs, abstraction, and taste rather than literal avatars. The self becomes:

• aesthetic affiliation

• intellectual posture

• system participation

• cultural signaling

A Squiggle can become a face. A Fidenza can become a personality. A Terraform can become worldview. This is where: identity becomes conceptual.

“A Squiggle can become a face. A Fidenza can become a personality.”

5 - IDENTITY AS COMPOSITION

This evolution is important and under discussed... But extremely important. Because the single PFP is insufficient for nuanced identity signalling. So more people begin layering. This is:

• PFP + banner

• PFP + art background

• PFP + reply style

• PFP + emoji language

• PFP + chain affiliation

• PFP + quote tweet behaviour

• PFP + wallet curation

• PFP + typography

• PFP + media aesthetic

In other words: identity becomes assemblage.

“Identity becomes assemblage.”

6 - THE ATMOSPHERIC FACE

Milady and the rupture. Milady was not simply another PFP collection. It represented a complete mutation in how identity behaved online. The image itself became secondary to circulation: posting style, screenshots, reposting, fashion, irony, emotional ambiguity, low fidelity compression, memes, photographed screens, timelines, recursive degradation. Identity stopped existing primarily inside the artwork. It became atmospheric. Milady collapsed:

• avatar

• posting

• meme

• fashion

• vibe

• irony

• subculture into a single contagious aesthetic field.

The repost became part of the artwork. Compression became texture. Low quality became authenticity. The face no longer needed stability. It survived through circulation.

"The repost became part of the artwork. Compression became texture."

7 - THE SCHISM

The earlier crypto-art world had largely stabilised around refinement. Even at its most experimental, Ethereum culture often still treated the artwork as a coherent object: intentional, complete, collectible, legible. But after Milady, something fractured. Part of the post-Milady underground no longer wanted merely to refine the tokenised image. They wanted to mutate it, inherit it, corrupt it, resurrect it, weaponise it against itself.

The movement toward Solana was not simply economic migration. It felt exilic. A drift away from an increasingly institutionalised Ethereum culture toward unstable new territory where images could evolve faster, stranger, uglier and more communally.

MiFella emerged first as both parody and infection, exposing the instability already latent inside the Milady form. DriFella pushed further still: no longer simply a derivative, but a resurrection mythology built from the remains of earlier identities. The derivative ceased behaving like imitation. It became succession. The original no longer occupied the unquestioned centre. The offshoots, mutations and corrupted descendants began generating their own gravity.

What emerged from this schism was not merely a new aesthetic but a new understanding of onchain identity itself: not portraiture, but lineage. Not stable selves, but memetic bloodlines. The face no longer represented who you were. It represented what cultural organism you belonged to.

“The derivative ceased behaving like imitation. It became succession.”

8 - THE BROKEN FACE

The portrait collapses into debris. By this stage the face no longer functions primarily as identity. It functions as residue. A compressed archive of moods, affiliations, reposts, memes, emotional states and aesthetic contamination. Projects like:

• MiFella

• MiFella 2

• DriFella

• LFVAD2K24

• Wickz

• Spiky DJ

• Drilady

• Chair

• Constant Fella

all push toward this condition differently. Some feel painterly. Some feel memetic. Some feel schizophrenic. Some feel almost spiritually corrupted. But they share one important characteristic: they reject clean symbolic readability.

The early PFP era wanted the avatar to be instantly legible. Punks: clear archetypes. Apes: clear personality branding. Even Milady retained recognisable coherence beneath the atmosphere. But now the face becomes unstable. Features melt into noise. Eyes drift. Masks overlap. Emotions contradict each other. The image appears attacked by the network itself. The self is no longer singular. It becomes recursive.

This is where the Avant Gay scene fully emerges as its own cultural organism. Not as a genre. Not as a movement with rules. But as a loose emotional condition spreading across collections, posting styles, editing aesthetics, memes and interpersonal relationships. A culture of emotional excess. Irony collapsing into sincerity. Beauty collapsing into ugliness. Humour collapsing into despair. The face becomes a battlefield where all these contradictions coexist simultaneously.

“The self is no longer singular. It becomes recursive.”

9 - DAMAGE CONTROL

Long before the post-Milady schism fully emerged, parts of the Ethereum Fine Art Movement had already begun visualising a similar psychological condition. Not the stable symbolic face of early PFP culture... but the self under conditions of permanent network exposure.

XCOPY’s later works, particularly Damage Control, feel striking in retrospect.

Not because they predict the Solana underground directly, but because they arrive at a strangely adjacent emotional atmosphere... identity as contamination, repetition, degradation, emotional residue and recursive circulation.

The screaming volatility of earlier crypto art gives way to something flatter, more ambient and psychologically saturated. Collapse no longer appears as an event. It becomes environmental... And this is where the essay quietly loops back on itself. Because the emotional instability that later erupts through Milady, MiFella, DriFella, LFVAD2K24 and the broader Avant Gay underground was never entirely absent from onchain culture. It was already latent in the network... the pressure of visibility, financialisation, circulation, performance, reposting, identity collapse, infinite spectatorship. Different artists and ecosystems simply metabolised that pressure differently.

“Collapse no longer appears as an event. It becomes environmental.”

This was never a straight line. Not a clean evolution from one collection to another. Not a hierarchy of artistic superiority. Not Ethereum replaced by Solana. Not old culture dying so new culture could win. It was drift. Contamination. Pressure. Mutation. Recursion... A decade-long process through which online identity slowly detached itself from the stable human portrait and became something stranger... atmospheric, memetic, fragmented, financialised, networked and alive.

The blockchain did not simply create digital ownership. It created conditions under which the self itself could mutate. And somewhere between CryptoPunks and Avant Gay, between the symbolic face and the broken face, the profile picture stopped being merely an image of a person... It became a cultural organism.

“The profile picture stopped being merely an image of a person.
It became a cultural organism.”

As they spoke, my mind immediately went back to a comment I made to @notthreadguy months ago... "not token OR equity… tokenised equity"...

That distinction suddenly felt much larger to me than it did at the time. Because the moment tokens and equity begin converging into the same thing, the psychological architecture surrounding blockchain changes completely.

And my mind immediately drifted toward a question I don’t think enough people are asking yet...

what happens to the perception of onchain art once the chain itself becomes associated with real ownership infrastructure?

Not speculation. Not meme coins. Not internet casinos... Actual ownership.

One of the more compelling essays I’ve read recently on this broader topic came from Vangoya. The piece argues (convincingly, in my view) that the public narrative around NFTs being “dead” sits in direct contradiction to what institutions, museums, galleries, and serious collectors have actually been doing over the last four years.

The core of Vangoya’s argument is that onchain art is already entering art history. Not hypothetically. Actively.

The article points to:

• acquisitions by MoMA, LACMA, Centre Pompidou, ICA Miami, and the Whitney,

• the embrace of generative and blockchain native artists by blue chip galleries,

• the conceptual lineage between onchain art and conceptual art movements,

• and the importance of blockchain provenance as a fundamentally new kind of property system for digital culture.

I broadly agree with the thesis...

But listening to the discussion around Bullish acquiring Equiniti made me realise I was thinking about the legitimacy question from a slightly different angle.

Vangoya’s argument is primarily cultural... art institutions are legitimising onchain art. While I agree that institutional participation is important for canonisation. I feel a little differently about legitimisation...

What I’m increasingly interested in is something slightly different... what happens when financial infrastructure legitimises the chain itself? Because much of the resistance to NFTs has never really been about aesthetics. It has been about ontology. People struggle with the idea that tokens represent anything “real.” Eventually, almost every critique of NFTs collapses into the same underlying sentiment:

• it’s just a JPEG,

• it’s disconnected from reality,

• it’s speculative abstraction,

• it’s internet-native but economically hollow.

But transfer agents occupy an entirely different category of reality. They are not speculative. They are part of the legal machinery of ownership itself. Companies like Equiniti sit beneath the visible layers of public markets handling shareholder registries, dividends, corporate actions, and ownership records for public companies. They are part of the infrastructure that makes equity ownership function at global scale. So when a crypto native company acquires a transfer agent, the significance extends beyond “crypto adoption.” It hints at a future where blockchain rails merge with the infrastructure of actual ownership. And if that process continues... if equities, debt instruments, treasury products, and eventually credit markets begin existing natively onchain... then the chain itself may undergo a dramatic perceptual transition.

It stops looking like a speculative environment containing disconnected tokens, and starts looking like a global ownership and settlement layer. That shift has significance far beyond finance. Because once people become accustomed to:

• tokenised equities,

• tokenised dollars,

• tokenised bonds,

• tokenised ownership records,

• programmable financial assets,

the psychological barrier around tokens weakens dramatically. A token stops being perceived as “crypto.” It simply becomes a digitally native ownership object.

And once that mental shift occurs, onchain culture benefits almost automatically.

The leap from:

“I own tokenised equity” to “I own a culturally meaningful onchain object”

is far smaller than the leap from

“I own dollars in a bank” to “I own a JPEG on Ethereum.”

The internet itself underwent a similar transformation...

Early internet culture was dismissed as unserious and detached from “real life.” Forums, avatars, pseudonyms, and digital communities were treated as peripheral until commerce, work, communication, and finance moved online. Once the infrastructure became economically indispensable, digital culture inherited legitimacy from the substrate beneath it. I suspect blockchains may follow a similar path. And if that happens, the significance of onchain art may no longer rest solely on museums, galleries, or auction houses validating it from above. Instead, legitimacy may emerge from below... from the normalisation of blockchain based ownership itself.

To the degens reading this (a cultural oxymoron!??!)... this won’t pump your bags tomorrow. These shifts rarely feel dramatic in real time. Infrastructure changes slowly. Perception changes even slower... But directionally, something important may already be happening. Because if blockchains evolve from speculative environments into globally recognised ownership rails, then the cultural objects native to those environments may eventually be viewed very differently... Not as random internet artefacts from a failed mania… but as some of the earliest cultural objects created inside a new digital property system the world slowly migrated toward without fully realising it.

The Through Tubes, We Now Are pieces I've collected don't look like finished things. They look like objects mid-discovery... forms arrived at rather than designed. Each rendered through thousands of line-strokes... patient and dense... the surface built up one quiet breath at a time. qubibi describes the outputs as looking like "weathered lumps of something." He's not wrong. But there is a precision inside that lumpiness that I find quietly radical.

What draws me first is the system. qubibi built the tool himself. Authored the instrument, then draws with it. The constraint is self-imposed from the bottom up... not rules placed on a process, but a process generating its own rules as it learns to speak. This is the same quality that pulls me toward Terraforms, toward Human Resources, toward artists like Waffles. The system as dwelling place, not just output.

The first piece I collected was from qubibi's own Manifold contract, one of the first four released. I wanted in before the vocabulary was settled. What I own is a form that floats like a landscape seen from altitude ... or a wing, a page caught mid-turn ... with a small constellation in the lower left corner, like a signature the system didn't know it was leaving.

From SuperRare I collected two more (one animated, one interactive) to tell a story. Not about ownership. About process. Three states of the same investigation: early draft; system breathing; system asking for your hands.

What makes this work sit differently from most generative art is the instability at its centre. The system doesn't obediently produce... it continues... and what continues exceeds what was initiated. The tubes outrun the hand. The figure lags. Sound arrives as a consequence of gesture, not a decision. This sits closer to Terraforms or Human Resources than to conventional NFT image-making... a persistent computational environment where participation reveals the gap between intention and system response. That gap is the work. The interactive piece wasn't optional... it's where the instability stops being observed and becomes felt.

The NFT space is heavily weighted toward North America. qubibi is a Japanese artist from Ikebukuro who built his hello world algorithm in 2010 and has worked inside it for sixteen years, recognised at Cannes and D&AD, largely invisible to the Anglophone market. When I find non-Western artists of this calibre working in this space, attention is the minimum I can offer.

Then there's the sound. The project's own description is specific. Sound didn't come first... it emerged. Lines generated sound; sound grew toward music. qubibi calls the tool perhaps a kind of instrument... something played, not just operated. Sound here isn't decoration, it's evidence that the system has acquired a body. When an artist treats sound as structural rather than atmospheric, the work asks more of you. It doesn't just want to be seen. It wants to be inside the room with you.

qubibi's underlying question (which parts of my creative process should involve my own hands) is the question underneath all of this. His answer is this series... everything. He builds the system himself and draws through it. The line is his. The form is negotiated. The sound arrives as a consequence of the gesture.

That's what I collected. Not objects. A negotiation, documented in three states.

Sitting with the exhibition after reading his text, I found myself pulled in a slightly different direction.

Less toward how these images resist the system... and more toward what they become after prolonged exposure to it.

There was a time when a damaged image signalled failure. Compression artefacts, jpeg noise, deep frying, bad crops, overlaid text, meme debris... these were signs of corruption. Evidence that an image had drifted from some intact original.

That distinction feels difficult to sustain now. Images no longer degrade after circulation. Circulation is one of the conditions under which they are made. They emerge anticipating reposting, theft, remix, screenshots, compression, speculation, fandom, irony... They are born already touched.... by platforms, by markets, by strangers, by repetition.

Perhaps the “bad image” is not a failed image at all. Perhaps it is the image after too much contact.

Where one reading of the “bad image” frames it as resistance (a way of breaking legibility, frustrating the algorithm, adding friction back into optimised systems) the works in New Bad Image often feel less like acts of opposition and more like consequences.

These images are not trying to escape the system. They are what remains when the system has already run too long, too hard, and too often.

Not refusal... Aftermath.

The exhibition’s framing self-made mythmaking is the key.

Because myth here is not stable. It is argued into existence. Improvised from fragments. Partially remembered. Misremembered. Contested in public.

In one exchange, the very idea of “Avant Gay” as a coherent scene is dismissed outright... something that doesn’t meaningfully exist beyond a loose influx of artists, traders, and platforms.

What follows isn’t a rebuttal. It’s an accumulation. Fragments surface. Early experiments. Discord servers. Patronage structures. Arguments on the timeline. Collections born out of conflict. Scenes forming and dissolving around money, attention, shared references.

“Read this whole account as part of the art.”

The instruction is not metaphorical. The scene does not exist as a stable category. It exists as a network of references, gestures, relationships, recurring images... something you trace rather than define.

Elsewhere, the same voice pushes further:

“There is no longer any option in this world but total and radical engagement… no new products but the self… a new kind of fella discovered every day at the crossroads of unlikely connections and contradictory values.”

And then:

“It’s about love, not lore.”

But that feels incomplete.

Because without lore, love collapses into something sealed... a closed loop accessible only to those who were there. A nostalgia trap.

And without love, lore calcifies into empty structure... symbols without affect, endlessly repeatable, easily captured.

The tension between the two is not incidental. It is the engine. Myth, here, is not a fixed narrative. It is a process of selective revelation... Breadcrumbs... Enough context for those willing to engage, but never enough to become fully legible.

Andreas Gysin’s announcement design quietly establishes this logic before the works even begin.

Typography disperses into particulate drift. Language fragments into signal and noise. Meaning dissolves, but pattern persists. It reads less like graphic design and more like a thesis image for the show: that coherence is no longer given ... it is something assembled through fragments.

Parker Ito’s Fashion Terrorist sits close to the centre of this field.

These works don’t simply reference contamination... they operate as contaminated surfaces. Floral romanticism, scanner textures, degraded luxury imagery, collectible ephemera and digital residue collapse into one another until painting itself begins to behave like circulation.

They feel less like images and more like objects that have been handled too many times. Damaged, but not diminished.

Petra Cortright sharpens this lineage.

Her presence matters because she established, early, that software native image making (default brushes, desktop aesthetics, digital layering) could carry the weight of painting. Here, her return to degraded webcam systems shifts the focus from interface to interference. Not the tool, but the breakdown of the tool.

The image clouded. Unreadable. Protected.

But the exhibition becomes more volatile when it moves into the Solana avant scene.

Because MiFella, Luhhfella, Evil Biscuit, Tojiba and Terrorism are not simply extending post internet strategies. They are mutating them into something closer to crypto folk cosmology. Here, derivative imagery becomes a generative engine rather than a limitation... Lore becomes medium. Collectibles become carriers of meaning. And perhaps most importantly... bad taste becomes a form of intelligence.

MiFella’s Deep Fried Milady is almost brutally simple.

An image compressed to the point of collapse. Hair, eyes, face reduced to saturated striations... a PFP pushed so far through circulation it begins to resemble a glyph.

Not decline. Pressure. An image cooked until it holds the memory of every transformation it has passed through.

Luhhfella’s stills move in the opposite direction.

Where the PFP smooths expression into legibility, these works reintroduce friction through gesture... scribbled, strained, excessive. The hand reappears, but not as mastery. As effort. As failure. As insistence. A refusal to resolve.

Evil Biscuit’s Blessings and Curse pushes toward full cosmology.

Layered animations unfold like a corrupted game system. Sprites, devotional imagery, and painterly backdrops collapsing into one another. It feels theatrical, occult, unstable. Less like a single image and more like a world trying to hold itself together. Not illustration... Environment.

Tojiba operates differently again.

Cooler. More detached. Corporate logic turned into aesthetic material. Branding as performance. Tool building as authorship. The “brand manager” persona functioning less as joke than as structure... exposing how image production, distribution, and identity collapse into one another.

Terrorism’s card works introduce something the show quietly depends on.

Tenderness.

Where much of the exhibition leans into overload, distortion and irony, these works retain a romantic register. Trading card structures suggest activation (power, effect, meaning) but the imagery feels drained, filtered, held at a distance. Potential without release... And yet, something persists. A sense that even within systems that flatten meaning, images can still carry feeling.

Ann Hirsch and Maya Man’s New Year, New Bitches sits in a similar tension.

Synthetic bodies. Generated faces. AI assisted undressing. The aesthetics are familiar (influencer, doll, calendar) but pushed beyond coherence. Femininity rendered as system, then broken through repetition and excess. Not critique from the outside. Collapse from within.

Across these works, a pattern emerges.

Not simply degradation.... Accumulation.

Each image feels like it carries the weight of its own circulation... the marks of being shared, compressed, reinterpreted, recontextualised.

Touched too many times.

There is, however, a risk embedded in this language.

Once degradation becomes recognisable as style... once the “bad image” itself becomes legible... it risks re entering the very systems it set out to disrupt. Ugliness circulates. Distortion stabilises. Even failure becomes aesthetic.

The question then becomes... can contamination remain insurgent once it is expected? The exhibition doesn’t resolve this... But it presses on it... And that pressure gives the work weight.

Because what New Bad Image ultimately reveals is not simply how images break... But how meaning survives after they do.

In an environment defined by infinite feeds, synthetic abundance, attention extracted at scale... authorship begins to look porous. Identity becomes assembled. And authenticity, if it survives at all, survives not as purity, but as residue.

As distortion. As scar tissue. As myth built from fragments.

The bad image, then, is not the failed image of our time.

It is the image after too much contact.

And it may be the only kind of image we have left.

It assigns you a role inside a system that already exists.

At first, this looks like an evolution of the PFP model. A move from surface to structure... From aesthetics to function.

But that’s not quite right.

Human Resources doesn’t just improve the PFP.

It inverts it.

Most PFPs are built on a simple premise:

Identity is something you project.

You choose an avatar because it looks like you, or like who you want to be. Traits signal taste, status, affiliation. The character itself is often a cipher. Story and personality are layered on afterward.

This model worked. It scaled. But structurally, it’s shallow.

Remove the social layer, and many PFPs collapse into what they always were:

costumes without a script.

That’s "The Flatness Problem".

Human Resources breaks this at the root.

It doesn’t start with identity.

It starts with a system.

A corporate structure. A taxonomy of roles. A set of power relations that exist before any holder arrives.

You are not inventing a character.

You are stepping into a position.

Traits are no longer aesthetic. They are functional. A compliance officer, a crisis actor, a false flag operative... these are not visual skins. They are roles that imply behaviour, constraints, relationships.

Identity is not expressed.

It is assigned.

And that has a consequence.

If identity is assigned by the system, then the holder is no longer its author.

You don’t decide who you are.

The system does.

Crucially, the system is not presented as broken. TabCorp absorbs everything—legitimate roles and conspiratorial ones alike—into a single operational logic. Even anomalies are incorporated.

Nothing escapes classification.

The avatar is no longer an extension of the self.

It is an extension of the system.

At this point, the format begins to invert.

Human Resources still looks like a PFP collection. It sits in the same interface. It can still be worn, traded, displayed.

But structurally, it behaves differently.

A typical PFP enables identity expression.

Human Resources constrains it.

A typical PFP asks:

who do you want to be?

Human Resources asks:

where do you fit?

This is where it becomes something else.

An anti-PFP...

An anti-PFP doesn’t reject the format visually.

It rejects it conceptually.

Instead of enabling self expression, it limits it.

Instead of amplifying individuality, it standardises it.

Instead of offering a blank canvas, it imposes structure.

Instead of allowing identity to be chosen, it assigns identity as a function of a system.

It doesn’t remove the PFP.

It inverts it.

But the inversion doesn’t stop at structure.

It extends into behaviour.

Through the daily prompts shared by @Taborp_HR, the system activates.

Each scenario... absurd, corporate, surreal... invites a response. And each time, holders respond not as themselves, but as their assigned roles.

What looks like engagement is something more precise.

The collection is being performed.

Roles stop being descriptive.

They become behavioural.

And through repetition, something shifts.

Each prompt reinforces the frame...

Each response deepens it...

What begins as play starts to resemble conditioning.

Not enforced. Not explicit.

But structured enough that participation begins to shape behaviour.

The system doesn’t just assign identity.

It trains it.

This is why participation matters.

Without it, the collection is static.

With it, the system comes alive.

Identity is no longer just metadata attached to an image.

It becomes something enacted, repeated, stabilised.

The collection is not completed at mint.

It is completed through use.

This is where the contradiction becomes real.

You are using the PFP to express yourself.

While simultaneously submitting to the identity it assigns you.

You are:

• wearing the avatar

• performing the role

• reinforcing the system

All at once.

The anti-PFP is no longer theoretical.

It is lived.

Most PFPs stop at ownership.

Human Resources extends into behaviour.

Not just: what do you own?

But: how do you act within the system you’ve entered?

The flatness problem was never just about aesthetics.

It was about authorship.

Who defines identity... the holder, or the system?

Most PFPs answered that question in favour of the holder.

Human Resources doesn’t.

It assigns identity.

It activates it.

And through repetition, it begins to shape it.

At that point, the question is no longer:

who do you want to be?

Or even:

where do you fit?

It becomes:

what is this system making you become?

That's what this collection does if you let it.

Most PFPs are collected as surfaces. Things you wear, things that map cleanly onto a version of yourself. But that breaks down quickly here. These aren’t avatars in the usual sense. They don’t resolve into identity. They resist it.

I’ve written before about how most PFPs struggle to hold narrative weight... they look complete, but don’t actually contain anything beneath the surface.

Every one of TabCorp's 10,000 operatives is, on paper, the same thing...  which is exactly why trying to “find yourself” in one of them feels slightly off: a standardised white male mannequin base with procedurally applied variations. The system is democratic in its indifference. But within that uniformity, certain classifications break from the standard operative model in ways that aren't purely aesthetic. The Clones, the Clowns, the Kill Bots, the Disappeared Staff, the Mannequins, the Gang Stalkers — these groups are doing specific ontological work inside the world Tabor has built. Each one represents a different relationship between the individual and the system that contains them. Each one is, in its own way, a critique.

The Clone: Bureaucracy as Existential Resolution

The Clone's tracking badge is not a security measure. It is a disclosure requirement.

TabCorp has looked at the situation... two near identical operatives, same anxious face, same hazmat suit... and made a determination. One will continue in their assigned role. The other will be assigned the role of Clone. Not spy, not replacement, not threat. Clone. As in: that is the job now. Report to your desk.

What this means is that TabCorp has bureaucratised the existential crisis completely out of the equation. The question of which one is real... the question that should shatter everything... has been resolved not by answering it but by making it operationally irrelevant. The badge tells colleagues what they're dealing with. The original continues uninterrupted, or continues knowing, which TabCorp has determined is not a distinction that requires action.

The Clone's critique of the system is this: the system is so total that even the proof of its own artificiality can be absorbed, classified, and put to work. The Clone doesn't escape the machine by being evidence of the machine's constructedness. It just gets a new job title. The badge on the forehead isn't a scarlet letter. It's a filing convention.

The Jailbroken Clone (#414) and its Jailbroken counterpart (#258) complicate this further. Two versions of the same operative, both outside normal parameters, both still running. TabCorp hasn't resolved the duplication because doing so would require determining which one is the original, and that determination lives in a form no one has located. So both continue. The system's response to an unresolvable question is not to resolve it. It is to keep both instances active and update the incident log.

The Birthday-Grade Comedy Core: The Tragedy of the Pre-Named

The Clown is the oldest archetype for the person whose job is to perform an emotion they don't feel. Which makes them a hyperbolic mirror for every other operative, who is also performing a role. But what distinguishes the Birthday-Grade Comedy Core from that general condition is the qualifier: Birthday-Grade. Not dangerous. Not transgressive. Not even particularly funny. Domesticated. Assigned to a corporate occasion. Rarity-controlled at 88.

Every other operative has some gap between their assigned function and their interior life. That gap is where the humanity lives, where the stories happen. The Clown doesn't have that gap, or rather: the system has already named what's in it. The role is the performance of joy. There is no inner life the classification hasn't already accounted for. The system has looked at the Clown and said: we know what you are, we've quantified it, we've assigned it a rarity score.

The Birthday-Grade Comedy Core's critique of the system is the most suffocating of all: what happens when the system has already named your deviation? Every other operative can at least experience the private ache of being misclassified. The Clown can't even have that. The comedy has been graded. The birthday has been scheduled. Show up and perform.

The Kill Bot: Violence as Compliance Theatre

There are only 24 of them. That number is the point.

If the Kill Bots were the enforcement mechanism (the means by which TabCorp maintains order) there would be far more of them. 24 is not a security force. It is a proof of concept. A deterrent kept just visible enough to function as one. The Legally Distinct qualifier is TabCorp's relationship to violence made linguistically explicit: not denial, but indemnification. We are not saying these aren't kill bots. We are saying that, in the relevant jurisdictional sense, they are technically not, in a way that satisfies the applicable compliance framework.

The Kill Bot's critique of the system is structural rather than personal: the system doesn't need to deploy violence often. It needs you to know the capacity exists. The 24 are less a standing army than a standing memo. Their rarity is the message. If you needed them frequently, something in the hierarchy had already failed. Their near-absence is the proof that the hierarchy is working.

The Disappeared Staff: The Glitch That Didn't Get Filed

"Abducted mid-shift, wearing whatever they had on."

No special wardrobe. No reclassification. No termination paperwork, which in TabCorp is significant because termination is not exit... it is repurposing, and repurposing requires documentation. The Disappeared Staff have been caught in a moment the system skipped over without completing the administrative cycle. They're still technically on the org chart somewhere. Their desk is still in the headcount. Their access badge still works, probably, somewhere, in some system that hasn't been updated because updating it would require completing the form that was never filed because they disappeared before it could be initiated.

The horror of the Disappeared Staff is not the disappearance. It is the incompleteness. Every other operative, no matter what has happened to them, has been processed. Classified. Assigned. The Disappeared Staff have been processed halfway and left in a liminal state TabCorp hasn't resolved... not because it can't, but because resolution requires a triggering event, and the triggering event was the disappearance, and the disappearance didn't generate the right kind of paperwork.

Their critique of the system is the most inadvertent: they have found the only exit, and the exit is a bureaucratic error. They didn't escape the machine. They fell through a gap the machine forgot to close. Whether that's better or worse than staying is not a question the system is equipped to process.

The Mannequin: Optimisation's Logical Endpoint

The Mannequin is a static display unit sometimes mistaken for staff. The inversion of the Clone problem: where the Clone is an operative who might be a copy, the Mannequin is a prop that might be a person. TabCorp cannot reliably tell the difference... or more precisely, doesn't need to. If a Mannequin is performing its function... occupying a desk, filling a headcount, maintaining the visual grammar of a staffed office... it is, by TabCorp's metrics, delivering against its performance indicators.

This is the endpoint of the optimising humans into resources thesis. At full optimisation, the human and the prop become functionally interchangeable. The Mannequin's critique isn't angry. It isn't even sad. It's simply true: the system was never measuring humanity. It was measuring the appearance of humanity. Once you can manufacture the appearance without the original, the original is redundant. The Mannequin is what happens when the KPIs have been perfectly met.

The Gang Stalker: Surveillance as Collegiality

No unique wardrobe. They blend by copying lower-tier clothing specifically... not management, not the elite classifications. They are embedded in the expendable classes, watching from inside.

The Gang Stalker is the system's immune response made flesh. Not enforcement (that's the Kill Bot's jurisdiction) and not management (that requires a wardrobe). The Gang Stalker is ambient. Present without being identifiable. Watching without watching from above. The fact that they copy lower-tier clothing is the detail that matters most: the surveillance isn't directed downward from authority. It runs laterally, through peer relationships, through the colleague at the next desk, through the person who seems to be having the same bad day you're having.

Their critique of the system is the most intimate: you are not being watched by the system from outside. You are being watched by the system from inside the same experience of being inside the system.

None of this is particularly subtle once you start looking at it this way.

The friction comes from somewhere else.

Most people are still approaching this like a 2021 PFP collection... looking for something to wear, something that reflects back a coherent, aspirational version of themselves.

But Human Resources doesn’t offer that. It assigns roles. It distributes positions inside a system that doesn’t particularly care whether those positions feel like you.

The watcher is indistinguishable from the watched. That's not a flaw in the surveillance architecture. That is the surveillance architecture.

What They Share

Each of these groups represents a different answer to the same question: what happens when the process of converting a human into a corporate asset goes wrong, goes too far, or gets interrupted?

The Clone is duplicated past the point where selfhood is a useful administrative category. The Clown is converted so completely that even their deviation has been classified and assigned a rarity score. The Kill Bot is converted into pure function, with the human substrate reduced to a compliance footnote. The Disappeared Staff is caught mid-conversion, stuck between states. The Mannequin was never converted at all but is indistinguishable from something that was. The Gang Stalker is converted into a surveillance instrument that operates through the performance of not being one.

Together they form a complete taxonomy of institutional capture... arranged not by what was done to them, but by which part of the process broke, and what the break reveals about the process itself.

The system is never disrupted by any of them. The Clone gets a badge and a job title. The Clown gets a rarity score. The Kill Bot gets a legal disclaimer. The Disappeared Staff get an incomplete form. The Mannequin gets a performance review it will never fail. The Gang Stalker gets a desk near yours.

TabCorp absorbs them all. That's the point... That's always been the point.