Artificial assistance

Late at night, one fateful Sunday last November, I tried ChatGPT.

The following morning, before a small undergraduate class, I presented it. On the one hand, I was acting like the father that hands a beer to his thirteen-year-old after a heavy meal — “I'm the one watching you drink your first beer” — the professor watching students' first experiments with a tool that has obvious plagiarism potential. On the other hand, I said something along the lines of “if you let yourself become the kind of worker who always relies on these tools, why should anyone employ you then?”, and emphasized the amount of knowledge and skill needed for vetting the system's output.

Since then, I've been meaning to write about Artificial Assistance systems, a term I still feel as more apt than the 'intelligence' moniker. Still struck from that late Sunday experience, I wrote a few paragraphs for a Portuguese magazine, concerned with the quality of the training sets and with what happens when the systems are trained on their own output. And I've read, read about A.I. and ChatGPT and Bard and Bing and GPT-4, as I read about and experimented with Midjourney, Stable Diffusion and DALL-E before. I began struggling under the saturation wrought by this always-connected age, paralyzed by the awareness that whatever I may think about something, surely someone else already thought it through and articulated it better. Why should I write (or query ChatGPT for assistance) when there is somewhere out there that article that I can just toot with a “This 👇 ?”

Computer History by Balenciaga, by Azlen Elza. I find it weirdly appropriate to see the grandparents of today's digital technology clothed in what it seems like the ultimate Veblen brand.

My contribution may be slim at best, if read at all. Nonetheless, here it is.

Most theories of technology acceptance emphasize Perceived Usefulness (sometimes named Performance Expectancy) as a key driver of an intention to use that technology. I am not aware of TAM studies about ChatGPT and the more recent language models, but I am sure their publication is being fast-tracked as I write. Nonetheless, on a purely anecdotal level, I've witnessed or been part to enough café conversations with people who, though not usually early adopters, have been querying ChatGPT for assisting in writing formal emails, social media posts, suggesting pub quiz questions, and a plethora of other writing tasks (and mind you, I don't hang around engineers' watering holes; these conversations took place mostly among arts or media people of some sort).

I am yet to catch a student committing A.I. plagiarism — and I've tried the tools — but that does not mean I haven't been deceived yet. All of this to say: I am positive ChatGPT's Perceived Usefulness measures through the roof, even among those people who I know not to be so usually enthusiastic about the latest technological thing. Moreover, another usual driver of technology usage intention is something called Subjective Norm, which can be construed as peer pressure. If everyone is talking about how useful ChatGPT has been, well... more people will use it. And indeed, there is no indication that one should doubt the assertion that ChatGPT is the fastest-growing product in history.

So this has been an 'iPhone moment' for artificial assistants: they've been around for a while (from old Eliza to Siri to Alexa), but ChatGPT presented an elegant package, highly useful and usable. Whereas before we struggled to get Google Assistant to 'understand' we wanted it to open Maps and show the best way to beat the traffic, now we had a humble text prompt take us nearer the Star Trek dream of a computer we can just talk to. Pandora's Box has been opened, or more like destroyed in a fit of difficult-packaging rage. Luddites notwithstanding, there is no going back to before November 2022: destroying the supercomputers that hosted the Large Language Models' training may prevent said training for a while, but the models themselves are out. A mid-range laptop can run Stable Diffusion as long as it has a recent dedicated graphics card. You can run Meta's LLaMA language model on your own computer as well, and its usefulness only bound by the amount of RAM (the better training data requires a lot — for 2023). Microsoft is putting its GPT-based Copilots in everything, from VS Code to Office apps (plagiarize from the comfort of your Word document!), surely to get that sweet OpenAI return-on-investment, but possibly also to protect their cloud computing business before too many realize they can roll their own artificial assistants on their companies' intranet.

And people will use these tools; I will use them, have used them. I do not want to dismiss ChatGPT et al's potential to generate bullshit — it is all it does, actually; truthfulness is a coincidence. I am surprised that the spam/scam-o-calypse hasn't happened yet. Still, in this world that is saturated in organically-sourced bullshit, I can see that artificial assistants offer diminishing returns. Using these systems might even (gasps!) be instructive! Hence, who pays and chooses the feed that goes into the training machine matters greatly. It is hard for me not to read A.I. luminaries' recent calls for a freeze in system development as a hypocritical, 'sour grapes' situation from ChatGPT’s beaten competitors and some academics clinging to their publication windows or — much worse — as a cynical ploy to vest some kind of 'infallibility' to artificial assistants in order to justify layoffs and privatizations and an (even) greater precarization of work. Indeed, I am more concerned by the decisions of business agents than by any implicit threat in the tools as they exist nowadays.

Created with Stable Diffusion 1.5: Two chatbots chatting while sipping wine, outside a Mediterranean café, photo taken with a disposable film camera.

Two chatbots chatting while sipping wine, outside a Mediterranean café, photo taken with a disposable film camera; Stable Diffusion 1.5.

My personal metaphor for these Large Languages Models is immutable software brains. Colossal amounts of energy and data and human work (in 'reinforcement learning', 'alignment', etc.) are employed in building these black boxes containing billions of 'neurons,' except these neurons are as if etched on glass: the 'software brain' responds deterministically like any simpler circuit, without any ability to form new connections. Any randomness is engineered, as it is in computing in general. ChatGPT may seem to follow a conversation, but that is merely a neat user interface trick in which the 'token window' expands as you go along — i.e., the interaction 'history' is appended to the user input. Does it understand?

I don't believe it does. Not only because that makes the mistake of anthropomorphizing A.I., which is quite hard to avoid: these are systems engineered by humans, trained on human contents (so far), meant for human use. Rather, a write-once glass-etched model of a brain's long-term memory is, by design, unable to learn anything, and thus to know anything about cause and effect. And thus, to understand, which would require an ability to experiment. And without cause and effect, how can models do anything but bullshit, pardon me, hallucinate? Borretti's anecdote about Bing's Sydney pleading to the user just tell us said plea pattern exists somewhere in the training data: train a L.L.M. on a diet of murder mysteries, and you will get detective-bot and murder-bot.

Nonetheless, maybe we do need a moratorium on that next step in which models are trained in real-time and 'learn' from interactions. That is perhaps the seal that is better to stay unbroken; the short step from atomic bombs to thermonuclear, and how Skynet wakes up. Or maybe not: biological brains developed over millions of years, whereas engineering is hard and humans are not good enough at designing antifragile systems. Perhaps the human memes in the future artificial intelligences turns out to be too much of a handicap: instead of Skynet, we get Bender: a danger on par with stupid humans near the Red Button, which we have been, somehow, surviving. (Until the day...)

So, which way is it? I would say we are all, as a society, about to get f*cked by A.I. — or rather, by the people who are, yet again, about to blame us for being uncompetitive against computers. If you ever thought “Gee, this task is surely repetitive, I wish I could have a program to do it,” you are about to have your wish granted, alongside all its unexamined implications. Artificial assistance is great in its potential to lower the amount of bullshit jobs and tasks people undertake, either willingly or unwillingly. If only our bosses would let us get away with it.

Created with Stable Diffusion 1.5: Karl Marx debates with a robot, outside the UN building, still from a newscast.

Karl Marx debates with a robot, outside the UN building, still from a newscast; Stable Diffusion 1.5.

The question about what to do, therefore, is one of politics. Not only in regulating the potential true A.I. (for which I have not much hope, so my fingers are crossed for intractable technical problems) and in regulating the training process of our artificial assistance systems, but also in addressing the challenges presented by the use of such systems. Much like we shouldn't — and shouldn't have — taken the internet's cornucopia of endless 'content' as the flip-side to the erosion of public discourse and citizenship and social conviviality, we should not take the usefulness that drives the inevitability of artificial assistance use as the flip-side to further inequality and generalized stupidity.

Given the visibility surrounding A.I. art, I believe artists and critics will have a very important role in setting the agenda for the evolving relationship with artificial assistance, though I am not holding out hope for most of the current generation after the NFT embarrassments of 2021. (I see copyright violation as one of the least interesting problems posed by A.I.. It is disappointing to see so many discussions proceeding as if an artists' skill is just a means to price commercial transactions in a market. There are far more interesting — and, ultimately, important — questions about who craftsmanship is for; if that skill is put at the service of doing illustrations for dropshipping businesses' social media campaigns, then well...)

The productivity gains of artificial assistants should prompt us all to discuss and demand much shorter working hours, universal basic income, or a reorientation of the education system away from the rote and towards the critical and the creative. It’s time we get our fair share.

This is a video I put together from renderings of the final daily iterative Processing sketches I made during during last January (#Genuary, get it?), set to some music I made with Magix (cough) Music Maker (cough).

The sketches' source code is available at my Github.

Totally fungible tokens

What have we — the digital art pedagogues — done?

Little did I know how ideas of the metaverse and of digital art would evolve during the last eleven months. I did eventually stop logging on to Occupy White Walls. The last months of writing before turning my thesis in demanded so, and I didn't feel like coming back after (finally!) completing my doctorate. Maybe rightly so, as metaverses no longer sound fun.

Still, in hindsight, Occupy White Walls appears as the gentlest metaverse, a place where art 'collecting' is unbound from any predatory pseudo-scarcity and only limited by time and imagination. I did not see how quickly radical financialization would conquer and reduce most of the digital art landscape to a trading cards business after so many creators accepted a Faustian bargain to take part in cryptocurrency schemes. Does the saying go, "if it exists, there will be crypto assetization of it?"

The embrace of NFTs by a large part of the creative code community is perhaps a minor scandal in such a bleak year. It struck me deep, however. That oily smell of money and greed usually lessens my passions. Rather than diving back into the tinkering and hacking I was yearning to, I found myself repulsed by Processing and creative code since completing my thesis. The promise of digital literacies, of Englebart's augmented intellect, of the superperson on the cover of Ted Nelson's Dream Machines — they who know how to 'decode the Matrix' — all that took a backseat to a Get Rich Quick scheme. Thus, n00bs were led to to code combinatorial art while losing their savings to crypto fees in order to place an invisible hand underneath art influencers' Ether and Tezos portfolios, metaphorical and literal atmospheres be damned. Software may or may not eat the world, but pyramid schemes surely will.

A glitched-out JPEG

So, here's a JPEG. It's already distressed and damaged, like a brand-new pair of designer jeans. You are encouraged to right-click on it and save it.

Curating in the metaverse

People often tell me I am lucky to be writing my Ph.D. dissertation in the midst of a global pandemic. I understand they believe that writing a thesis requires one to self-quarantine, and thus expect me to be minding my methodology chapter oblivious to the raging global apocalypse outside. Yet, as with many well intentioned questions and comments regarding one's ongoing doctoral penance, I wish people would just stop. Indeed, there are parts in which a pandemic and intense, long, never-ending writing and editing mix: I don't get as many invitations for things that might derail my work, and when I do, people are more understanding when I decline. Fear of Missing Out is gone.

Nonetheless, those who know me better know that I'm not good at being a shut-in. If no coronavirus ever mutated and 2020 had been a 'normal' year, maybe I would be regreting the times I stayed out longer than I should, had one more drink than I should, that party and that barbecue that would be the reason I had not worked as much as I needed to on my dissertation. Yet, I would welcome the constant changes of scenery that always helped me being a productive citizen. Walking, driving, riding the subway, laptop on my backpack: writing in cafés, writing in academic libraries and academic bars throughout the city, writing by the beach in the summer (struggling with glare and polarized sunglasses that darken the screen), writing in hotel lobbies in the winter, writing in Porto, writing in Lisbon, writing in the patio of some guesthouse somewhere South I haven't been before — the whole TV commercial for mobile internet. Where I wouldn't write much, save the odd email, would be at home.

Yes, I have an okay desk and have been moving my computer back and forth to the living room table so I can change the scenery. I even bought this small table with the exact height that I can place on the dining table and write while standing in an ergonomically correct posture so I don't get fat from my Ph.D. as long as I hold the comfort food in check. (I'm writing while standing right now.) Yet, there are far too many distractions at home I had to learn to live with. In the old days of the first lockdown, when we didn't know whether there would be a vaccine or a cure, or whether it would rather be a countdown to a megadeath 'herd immunity', I went on a fuck it attitude for a month, playing Xbox and watching a lot of movies and reading comics and planning weekly walks of a few blocks around where I live with the intensity one plans a raid behind enemy lines. I also looked out of the window a lot.

After snapping out of it and getting back to work, I settled on a few habits — poorly scheduled, but habits nonetheless — that helped me frame the time I dedicated to writing. I signed up for a sports streaming service and got into German football and Formula 1 racing — sometimes a nice lively background noise in the weekends, sometimes appointment viewing. I stopped watching movies (too long) and instead I watched one good TV episode per day (watched all of The Sopranos, all of Deadwood eventually). I stopped reading anything not related to the Ph.D., even now that I'm just writing the damn thing up, because reading became an activity loaded with intent and guilt. (Looking forward for leisurely reading sometime in 2021!) And I started to spend some time in Occupy White Walls.

OWW

Occupy White Walls (henceforth OWW) is massive multiplayer building game, somewhat reminiscent of Minecraft and Second Life, where you create art exhibits. Big art exhibits. You have a massive collection of elements — floors, walls, ceilings — to place in a three-dimensional orthogonal space, which you can populate with furniture, light fixtures, and artworks. You buy 3D space, elements, and the artworks using the in-game coin, which — and this is key — you just earn automatically by setting your space open for visitors. No real money is involved, and there's no other manner of earning game coin. You pace yourself: you build, you reopen, you wait. You put the kettle on, go visit other players' spaces, or work on that dissertation. After a while, you have earned some coin so you can continue to build or buy reproductions of artworks to place on your walls. The in-game collection includes paintings from the Renaissance to the early 20th century (they have some kind of deal with the London National Gallery and the Birmingham Museum and Art Gallery) as well as artworks uploaded by players.

OWW

Artwork uploads are, in fact, the only thing OWW charges you money for. They have some kind of dealership thing going on (which I'm not interested in), but I find charging for uploads quite a clever business model: it severely moderates porn while paying for the game — you can download OWW and play it for free. Minding the graphics settings, it runs well enough on my laptop, which is old but also has discrete Nvidia graphics.

I like it that OWW has no objective but to build, collect, visit other players' spaces (each can build up to three), and take nice screenshots (there's an in-game 'drone' for that). Nobody keeps any sort of score. It's like this metaverse art garden you tend to and curate. You can go nuts and just design whatever impossible, AESTHETIC (there is a vaporwave shader you can apply), space you want. I found out I am a realist, trying to make buildings that look like they have structural (collumns and vaulted ceilings) and historical integrity: I started designing these small palaces or industrial warehouses using the older-styled building blocks, and then I threw these wacky postmodern architectural renovations at them. Perhaps because of their groundedness, I started to dream (actual sleeping dreams) of my OWW spaces... but interestingly, dreaming of the modifications I would then attempt on my next visit.

OWW

I know Occupy White Walls has been often a welcome respite from being locked in. Looking at my massive spaces, which take a while to load — search for Moore Museum or Moore Hill inside the game — they have made me appreaciate how a little tending every day goes so far, perhaps more, if I should say this on record, than the 250-plus pages in my dissertation docs so far, so full of placeholders and notices and yet unrevised text. As metaverses go, OWW is a very good one. Yet I long for a day soon in which all those screenshots among my photo backups merely signal a lockdown past.

The young seagull

Like most city-dwellers around these parts, I do not like seagulls very much. Their cries may be a part of the seaside charm while walking the boardwalk by the beach, but in this city seagulls take the place held by rats in many other cities' folklore. Stories abound of seagulls snatching someone's croissant (because Europe), seagulls feasting on the abandoned pizza boxes of some impromptu picnic, seagulls tearing pigeons' heads off mid-air (a grisly sight I can't unsee). Not to mention the usual divebombing poop attacks, for seagulls can, unlike pigeons, crap on your brand-new jacket as you leave your just-washed car while in mid-flight.

Yet, I came to grow I-would-not-exactly-say-fonder of seagulls during the coronavirus pandemic. During the quarantine days, I spent many lonely hours looking out of my balcony, and I came to appreciate seagulls' flight. Though I wouldn't describe it as majestic or anything of that sort, the alternatives consist of the mindless circular flight of pigeons I got to know so well from my youth, the manic teleportation of elusive thrushes, or, come spring, the chaotic jetting around of sparrows. By comparison, seagulls' flight patterns look ponderous and dignified, in a way beautiful, closer to my memories of griffons in the Northeast, or that menacing black dot of a hawk we had christened Radamel (a weak football pun) circling over our house during holidays in Andalusia.

Come late spring, things changed. Seagulls became slightly terrifying. For the pandemic took away all the tourists along with the croissants and the leftover pizza, but probably also the city Animal Services' special birdfeed laced with sterilizing medicine. Going out of home confinement, people noticed how aggressive the pigeons by the sidewalk cafés became, hungry and craving cigarettes (true story). What about their bully and sometimes predator? There is a seagull nest up on the roof above my upper-floor apartment. Stepping into the balcony became The Birds by Northwest. Seagulls would circle and dive at me, crying, launching their webbed feet in front as if they were talons, and then suddenly pitching over the roof at the last second. A threat, a game, probably both. I would go to the balcony to water my plants and see how long I could stand in place before the mock attacks became too frightening. Sometimes I would put my hands up, showing them my palms, as if seagulls could somehow understand human gestures. Once I attempted to make owl sounds, and if I did not know seagulls' ka-ka-kas were symptoms of distress I would think they were laughing at my foolishness. Not that the owl sounds made any difference. These were seagulls in despair. Probably too many eggs, then probably too many baby seagulls, and probably not enough croissants and pizza.

So today I woke up with seagull noises as usual. But then I heard a thunk! and a new, higher-pitched cry. I opened the shades and I saw this bird in the balcony, certainly larger than your average chicken, ugly brown feathers half-caked in mud, black beak and webbed feet, a head that looked fuller of hair than feathers. A bird I am certain I had seen in some painting of Hell or at least Purgatory by Hieronymus Bosch or Pieter Bruegel: a young seagull. It paced back and forth in the balcony. In fact, the young seagull paced back and forth in the balcony in this very particular way, as if making an effort to hide its impatience, that actually reminded me of the way I would pace back and forth in that precise place during the quarantine, checking my watch's step counter at every end for no purpose but to beat the boredom. I looked at the seagull for a while, behind the shades as it had seemed frightened to see me. It paced back and forth, and after a long while clumsily opened its wings and tried to take flight — and couldn’t do more than jump a few centimetres. The bird seemed frustrated and resumed pacing, annoyed.

It begun to dawn on me that the seagull was not going anywhere, and I couldn't put it back in the roof, for I didn't have a big enough ladder, nor the necessary courage to grab a wild bird and take it up with me. I texted a couple of friends that know more about birds than I do. One of them advised me to give the bird some water, so I put this cup of water outside, slowly opening the door as not to startle the bird. At this point I was afraid the seagull would nevertheless manage to jump over the railing and plunge to the street. I looked online for the city Animal Services, while the young seagull settled in a routine of pacing around, crying, pecking frantically at the door, attempting to fly. The only direct phone number I found was listed as "weekdays — 10am-4pm". So, even though I felt it was a weird recommendation, I followed the website’s instructions and called the cops.

The police came, a little later. I joked at the officer who came through the door that I was aware this was a bizarre call. He was cool and told me it was the second most bizarre call of the morning, as he had just come from this place where an automobile had crashed inside a patio with no obvious way in. Okay, then, you win. He came into the apartment, looked out to the balcony, and called this private Animal Services number. "This kind of thing has been happening all the time," and left. A few minutes later, this very tanned old guy came in one of those old APE-50 cargo motorcycles. He didn't have a mask or helmet or any other kind of coronavirus protection, but looked like the kind of person who was bitten by enough dangerous animals to be vaccinated already — the kind of man you envision exterminating rats or grabbing feral dogs by their collar with those long poles. He grabbed this transportation box, the kind you take cats to the vet in, and walked the stairs, muttering. “July and August, always this shit.” The man stepped out into the balcony and swiftly grabbed the seagull. Immediately every single seagull in the vicinity took flight and started crying, an infernal cry. The young bird tried to resist by opening its wings, but nevertheless the man shoved it inside the cat box. High above, the neighbourhood's seagulls cried frantically all afternoon.

I was told the young seagull is to be taken to this Bio Park, not far on the other side of the river, where it will be fed and eventually released once it is strong enough to fly. I wish I could tell the older seagulls this: "You just fly straight south for a few kilometres, there's this place with weird birds you probably haven't seen before, there are these pink ones called flamingos for instance, and your kid will be there. And do not worry, they sometimes have hawks and owls and eagles there but those are in cages, and they only release them much farther away." But later, when I went to water the plants and play a little chicken with the seagulls, there was this extreme viciousness to their approaches, and it was not just the parents but the aunts and uncles and older cousins too letting me know that if they had talons instead of webbed feet they would tear my face off. I tried to protest "but you left your kid in the balcony to die!" but these are seagulls, they cannot pass the mirror test, let alone understand counterfactual histories. I was reminded of Werner Herzog's comments regarding chickens:

Look into the eyes of a chicken and you will see real stupidity. It is a kind of bottomless stupidity, a fiendish stupidity. They are the most horrifying, cannibalistic and nightmarish creatures in the world.

Werner Herzog was born in Munich, Germany, where I imagine there are no seagulls. Though, in regard to the first link, I must admit that seagull hypnosis sounds way more far-fetched than chicken hypnosis.

I hope that young seagull is okay, though.