'Good' 'Art' — What to argue about AI on the internet

AI art is not art. But why?

My many dozen readers might find this difficult to believe, but I enjoy arguing with people online. Lately, most of my pointless internet conversations have come in the form of arguing with tech stans about AI. My position isn’t particularly novel; like a lot of people, I find the widespread uptake of AI dangerous, myopic, hubristic, profoundly anti-life and a little depressing. Unless these tools are taken out of private hands, the only future I see is a permanent, boring horror. Like Salo.

There’s a zillion things to worry about. Far from being the democratising force some claim it to be, we suddenly have a genuine problem with not knowing if anything we see is real, a benefit to none but the powerful. These powerful people seem to have thrown all their chips on the side of fascism, which has historically proven to not be very nice. Jobs are being slashed regardless of the efficacy of AI implementation, yet the technology exists in such a total hype that even if it fails now, they will claim it will succeed in the future, whether that’s true or not. Naturally, this had led some people to worry, others to panic. Some of the loudest panic is coming from the visual arts world. Being from a humanities background, this is the area I feel most comfortable fear-mongering about.

I’m going to go through some of the major issues with discussing AI art, and break down what the arguments are, where they’re futile, where they’re effective and where they’re not, and hopefully provide some avenues for people to explore. This is a long article, so you are mandated to read it in a single sitting. I’ve made my biases clear; this will not be a complimentary article towards AI artists, and it will be outright hostile towards AI companies. The well is more poison than water at this point, so conducting myself like some neutral debate moderator against such a malignant industry seems an exercise in submission. That said, in the process of writing this article, my position on the utility of what we call AI has softened slightly. So if anyone who is pro-AI makes it to the end, I promise I will do my best to be fair. Not neutral, not nice, but fair. This is a courtesy Silicon Valley has never, and will never, give you.

Good’s got nothing to do with it

Early in the emergence of widespread AI-generated art, the knock against it was largely due to its technical faults. Hands proved difficult, text proved difficult, faces often ended up looking like deformed war veterans. As time went on, the AI got ‘better’ and the art became ‘good.’ The thing is, AI art advocates consistently appear to only be able to engage with art on a technical level. Their arguments rest solely on the machine’s ability to simulate the mechanical aspects of human artistic expression. If it looks accomplished on a technical level, then the art is ‘good.’ Good art, however, not only changes with aesthetic tastes over time, but rests on more than just the methods used in its composition.

Take a look at this picture.

This is some fan art of the Song of Ice and Fire character Jon Connington by Filipe Ferriria. I’m using fan art for this article because it is exactly the kind of popular art that is at once in great danger of getting drowned out by AI slop, but also because it is a great example of what art means to people. Fan art cannot be copyrighted; you cannot make money off it. Fan art is born of love for something; people feel compelled to take some of the precious few minutes or hours they have on this earth and devote their hard-earned skills or God-given talents to expressing affection for a particular story, world or character. Fan art is also known for being extremely inconsistent in quality, and we’ll get to that.

First introduced in the largely moribund A Dance with Dragons, Jon Connington (or JonCon to the fandom) didn’t appear next to any of the breasts of Game of Thrones, but is a thoroughly interesting character nonetheless. He’s an ageing exiled lord, a former hand to the Mad King Aerys II, a respected mercenary leader, a schemer for the Iron Throne, and a lovelorn adoptive father. JonCon somehow managed to survive failing the Mad King, a 1-on-1 with Prime Robert Baratheon, and a decade and a half of sellsword work. I’d add that he’s an active and practising homosexual in a medieval world, but bro ain’t practising, he’s mastered it. JonCon is a character haunted by his failures: his failure to kill Robert, his failure to his house, his failure to the man he loved, Rhaegar Targaryen. He’s also on his way out, afflicted by an incurable disease that will slowly but surely render every one of Jon’s failures a permanent, irreparable damnation. With his legacy hanging from a thread, he plots to place Aegon, supposedly the son of his unrequited love Rhaegar, on the Iron Throne by any means necessary. Let’s take another look at that picture:

Did Filipe capture the man or what? The dipped head, the weathered features, the exhausted but determined expression. The streaks of grey through his red hair speak to both his mortality and his resilience. His mood is low, his head is bowed, but his stance is strong; nothing on earth will move him from his mission, not the forces of nature and certainly no mere man. You can hear the wind blow and the sea spray across his face as he beholds the lands he means to conquer. The dark clouds speak to both JonCon’s intentions and the unforeseen consequences of his insane plan to upend an entire government. A storm is brewing and Jon Connington is right at the centre of it.

Now, if you’ve had a good look at this picture, you might have noticed there are a few problems with it. This is no diss to Filipe, I obviously like the image, but it serves as a good way to begin my point about how art works and where its value lies.

Take a look at this hand:

Look at the way it's positioned, the way it grips the sword, the way it overlaps the crossguard. He’s not really holding that sword, and if he was, based on the orientation of his wrist, the sword would be either through his leg or behind it.

Now take a look at the torso:

It’s twisted. His lower body is twisted to his right and his upper body is twisted to his left. If someone is standing tall, the way JonCon appears to be in the full image, would they really position their body like that? Would you, as an artist, position their body like that if strength was what you wanted to convey? Combining a shot of the torso that allows you to see the full arm also really drives home the issue with the hand and the sword, their awkward position.

Lastly, the castle in the background:

Perspective is tough, but the size of the castle just doesn’t work. Either JonCon is twenty feet tall, or we are looking at Storm’s End in the form of a Lego castle. The placement of his left leg is awkward as well, almost moving the castle parallel to the person.

Now let’s keep these flaws in mind and zoom back out to the full picture:

How much of the meaning of this image is lost knowing that there are technical faults in this image? The answer is none. The exact same ideas are conveyed about Jon Connington. Even if I hadn’t given you that preamble about his backstory, you’d still be able to tell that this is some sort of ageing soldier-man on some sort of final quest. To engage with art isn’t to get bogged down in technicalities. In fact, Filipe’s strengths as an artist are highlighted once his errors are acknowledged.

His work on Connington’s face is virtually flawless:

Such a specific emotion, such a specific expression. The shadows and lines on his face, the muscles underneath his skin, the placement of his grey hairs, the unique receding of his hairline tell us so, so much about the character. This is the sort of visual writers are discouraged from going into detail about because to get there you’d end up wasting a page writing about the placement of his eyebrows when you could just say “Jon Connington frowned” and rely on the surrounding context for the audience to create the frown in their mind. But an artist can, and should, draw this exact emotion. Human eyes are drawn to human faces, and Filipe knows this, so whatever else is going on in the image, the thesis statement is this man and what he is feeling. The eye notices one thing, the brain notices the rest. While you are drawn to the human, your brain processes the rest in broad strokes. This highlights his other strength, his use of colour. The image is dark but contains little black; the image is brown but is neither boring nor gross. Flashes of red connote aggression, the muddy backdrop warns of futility. You get all this information in a fraction of a second, without needing to be told who this man is and what he does. The many choices made by Filipe are acting all at once, informed by the source material, his own tastes, his own skills.

There are better educated people than me who can explain the triumphs and errors he made on a technical level, I just be saying shit. What matters is whether or not this image is ‘good’ is largely irrelevant to its status as art. At a certain point, when looking at any piece of art, a person will decide if they like it. They may eschew the big picture for the small details, they may, like me, prefer to take the image in as a whole, and appreciate the meaning conveyed by the artist. Yet there is no sensible debate or discussion to be had as to whether this image is art. It is, you know it is. There’s another absurdly long article I could write about what is and isn’t art, but the simple truth is, we know art when we see it; it’s something we figure out instinctively and there isn’t one solid answer. If there were, there wouldn’t be any debate on AI-generated art, we’d be able to point to one, two and three, and say yay or nay. All that’s really left is unpacking a series of abstractions. Feeling, value, meaning, the role of the artist in the creation of art.

Unfortunately, speaking in more abstract terms is often like trying to explain morality to a fire hydrant. When things can’t be broken down into a series of ones and zeroes, into a series of empirical data points, some people malfunction entirely, others refuse to engage, dismissing deeper analysis as pretentious, others fall back on empty solipsism: ‘It’s art because I like it.’ Why someone might prefer one piece of art to another is largely rolled up in their own tastes; someone may prefer an AI-generated image to a hand-drawn one, and they are free to do so. However, we do, whether anyone likes it or not, place certain creative outputs at a higher level than others. We value a clever novel over a clever tweet, a great album over a great advertising jingle, and we don’t need an argument to understand why the guy who thinks the McDonald’s theme is better than Abbey Road is worthy of minimal consideration. AI art is currently widely referred to as slop. The question is, is that a fair assessment, and why do people feel that way?

If the quality of the art is irrelevant to its definition, perhaps we should look at the role of the artist instead. Can an AI create a picture of Jon Connington? Probably. Will it be good? It doesn’t matter. Does it know what it’s doing? No, it factually does not. And neither does the prompter.

Yes, it matters how the sausage is made

When a person sits down to make a piece of art, they may use a pencil, a tablet, a paintbrush, a guitar, a camera or whatever. They draw, consciously or unconsciously, on what they know about creating a piece of art, skills and techniques that would best serve the piece they want to create. They’ll think about what they want to see, say, or what they want the audience to feel, or perhaps they’ll simply look to get something off their chest and let their instincts guide them. At every stage of a human creating a piece of art, a human is involved, from conception to creation, editing and refinement, to completion. This person is an artist, someone who creates art, good or ill.

When you ask an AI to generate an image, the process is a little less visceral. The AI is trained on countless images; these images are used by private companies for commercial purposes without consent or compensation, and they never will be. The AI reduces the images to a mathematical representation of said image, and a neural network makes connections between these data sets to learn what constitutes an image of a cat, or an image of a house or an image of former New Zealand pace bowler Tim Southee. There are a range of ways in which AIs can generate an image from there, but I’ll use the diffusion model for my example here because it’s the one I looked up. When you prompt an image, the AI reads your prompt and then gives that prompt its own mathematical representation, then applies that calculation to the various statistical correlations it established while being trained. It begins with random noise, then gradually refines it from there, using the pattern recognition it developed to create colour, shape and texture, until it spits out your image of former New Zealand pace bowler Tim Southee.

Now, you might say that both are using what they know to eventually create an image. The difference is that in the actual creation of an image using AI a human is never involved. Yes, they create a prompt: “I want a big titty anime girl with blue hair.” They might even feed the anime girl back into the AI to refine it, but at no stage is a human doing any of the actual work of creating the art. A machine is taking what you say and comparing it to what it knows and making a series of calculated guesses as to what you might mean. It does so without emotion, without motivation, and without an understanding of the context in which the image it generates comes from, or what it will exist in. There’s an inherent element of randomness, a loss of intent, that removes the human control over the image and over the message.

Having an array of skills and a base of knowledge that extends beyond the mere mechanical aspects of art creation allows a human artist to make choices to convey this intent at every step of the process. A pencil sketch may be preferable to an oil painting for creating a certain ambience (credit to Marc Fishman, I promise there won’t be more Song of Ice and Fire fan art), regardless of whether the oil painting requires more skill and effort. That is something an AI trained in simulating stylisic data does not know. It can know how to create a pencil sketch, but it does not know whether a pencil sketch or an oil painting would be the best way to present a piece. Only a human can know that. Crucially, it appears to be something that AI prompters refuse to know. There’s a consistent march towards attaining technical proficiency among AI art defenders. Sometimes great art is ugly:

It’s ugly because the circumstance it presents is ugly. Goya’s painting would be in no way improved by leaning more towards photorealism, more accurate proportions and anatomy. Don’t believe me? Check out Reubens’ far less interesting version. There’s a nightmarish trend of people posting their art online only to have some self-satisfied slopmonger ‘fix’ it with AI. There’s an element of online casual cruelty to this; there’s an element of the STEMlord superiority complex, showing some poor fool who’s made something real how they’re being ‘left behind’ in real time. I would also not be surprised if there’s a smaller contingent of naive people who actually think they’re helping. Regardless of why they’re doing it, the manner in which they do so speaks to a lack of understanding as to the role technique, medium and material play in artistic expression. Despite the so-called revolutionary tool they are using, they can still only engage with art as an audience, not as an artist. They nitpick, critique, and offer corrections so that they like it more, not so that the art can be better. They don’t know how to make art better; they don’t know what makes art better. They don’t know what they’re doing at all.

Someone prompting a piece of AI art is not an artist any more than someone ordering a pizza is an Italian chef. Any amount of detail in the prompt they use is like specifying which toppings you want on the pizza. But you did not make the pizza, you did not cook the pizza, you merely received the pizza. You can complain if you’re not happy with the pizza, send it back and ask for another pizza, but you will never have made the pizza. The meal itself is entirely out of your hands. You didn’t do anything. The pizza may be ‘good’, but it is not a creation of yours. You’re also limited by what the pizza place has available. You might want a pizza with, I dunno, penguin on it, but if they don’t have any penguin, you simply cannot get it. When you create art using AI, you are limited by what the AI has been trained on, you cannot create anything new, you are always going to be fenced in by twelve billion mathematical equations telling a machine what is, and is not, a shoe. You see disparate people separated by oceans and language creating comics and pictures constrained by the same Midjourney or ChatGPT house styles. The art isn’t theirs, it’s a particular machine’s version of what they want.

I’ve vibe coded up a few apps you can find on GitHub despite having no interest, talent or passion for coding. Do they work? I don’t know. Are they good? If the AI did what I told it to do, absolutely not. Do I know if it did? Nope. Do I even know how to take them from being code to being an app? Hell no. But therein lies the problem. I can use AI to create something I don’t understand, don’t know how to use, and don’t know what to do with. Am I a coder? I possess neither the knowledge nor the talent for coding. Is the AI a coder? Well, no, it’s not a person. Do I know what it did? Not at all. So is it really my code? No, it is not. If these apps exist independent of a developer, what are they? Sure, you can call them apps for lack of a better term, the same way we call AI art ‘art’, but we know there’s a difference between the two, that one is ‘real’ and one is something else, something mass produced, of dubious value, an afterbirth of RNG. The good news is, we already have a word for this: Slop. My apps are slop. AI art is slop. Neither will ever be anything else but slop. The common denominator between the two is that they were made using AI. They will always be slop, because they were made using AI.

Here’s another analogy: the fisherman and the bottom trawler. The fisherman goes out to the lake, with a rod, some bait and one of those hats with spare fish hooks in it. He casts out and catches a fish, maybe two, maybe more and then he goes home. Afterwards, he eats the fish. The bottom trawler works for Industrial Meat Production Inc. He presses a button, and a net the size of Vatican City scrapes the ocean floor, catching everything that falls into its grasp. Some of what the net catches is junk, some of it will be endangered species, and some of it is the fish that he meant to catch. He has no control over what falls into the net, though. He sorts through what has been collected without concern, based only on the criteria of what Industrial Meat Production Inc wanted to catch. Afterwards, he eats the fish. The man with the fishing rod is a fisherman, we know that. Is the bottom trawler also a fisherman? Both achieve the same goal: they catch fish, and people eat the fish. Yet we know there’s a difference between the two, that while both may be termed ‘fisherman’, they are engaging in entirely different activities. One has a skill for catching fish, and the other operates a piece of elaborate machinery. If bottom trawling sounds like diffusion to you, then congratulations, you still have command of your default mode network.

Socrates was wrong about writing, but that doesn’t mean AI is as good as literacy

A common thing you hear when facing accusations of Luddite-try is that anyone who has an issue with AI is just like so-and-so who thought blah-blah-blah would be the end of X. Socrates is a common one. Socrates thought reading and writing would be the end of people’s memories, as why would they retain any information if it was all stored in a book or parchment or CD or whatever they used back then. Socrates was wrong, yes, but is a man from 400BC who didn’t know about the invention of Mexico being pwned in 2026 really the mic drop it’s meant to be? He was a philosopher; he had an idea of how a person thinks, he didn’t know how the brain works, he couldn’t know how the brain works, they could barely deal with a broken leg back then. Besides, what’s more interesting is how he was wrong.

Reading and writing not only don’t damage your memory, it improves your brain’s neuroplasticity. Creating any sort of art is extremely good for people; even making crappy art helps form new neural pathways in the brain. Drawing, for instance, activates areas of your brain that are also responsible for verbal processing. Doodling while trying to figure out what you need to write next literally activates the part of your brain that’s active when you write. Any creative action is beneficial for more than just the intended task. Moreover, the manner in which the human brain and its over twelve different neurons are arranged means that certain parts of the brain are attuned to more than one task. It’s why you come up with ideas while walking to the shops. The parts of your brain that handle ‘walking’ and ‘shop’ may overlap with the parts that write “N.Y. State of Mind,” regardless of your intent to write a song or not. Think of all the brain activity that is not happening when you prompt your anime girl rather than reaching for the dreaded pencil; you’re doing yourself a rather holistic disservice. For more on this, I recommend the book Your Brain on Art by Susan Magsamen and Ivy Ross.

There aren’t a lot of studies out on the mental processes involved in generative AI use, and no meta-analysis that I can find, but what is out there isn’t pretty. A 2025 MIT study found that in academic essay writing, relying on AI to do the work for you was actively detrimental to the student’s learning, damaging their neural connectivity, cognitive processing, and memory. A Swiss study found that there are significant amounts of cognitive offloading occurring during AI use, especially with regard to critical thinking skills. Researchers out of Pakistan concluded that AI use is detrimental to memory and fosters psychological dependence. It may be that Socrates isn’t actually pwned in 2026; he was just talking about the wrong thing two millennia too soon. Actually, give these a read if you can. By 2040, they’ll be illegal.

Another common point of order made by defenders of AI art is the invention of photography. Visual artists had a minor freakout because cameras could capture actual reality, running Hans Holbein out of business. The long, steady increase in artistic fidelity to reality appeared to have been erased by a new technology overnight. However, we know the painter wasn’t eliminated by the camera. The invention of the camera instead led to bolder movements in the art world, one where capturing reality was eschewed for abstract, representational works:

Photography’s utility as an art form was not fully understood in the immediate seconds following its emergence. It became about capturing something real, a moment, a feeling, making something ephemeral permanent, preserving an instant as a monument as time rushes everything into the grave. AI can generate photos as well as it does anything else, yet much the way its art isn’t art, its photos aren’t photos, even less so. Cameras work by capturing real light reflected off real matter; nothing in an AI-generated photo is the result of any of that, no reality is captured in an AI-generated photo. And therein lies the problem with that comparison. The best-case scenario for a successful AI company is that its machine can do anything a human can; they have repeatedly announced this as an aim. A portrait artist can adapt to photography, offering things a camera cannot, a photographer still has a vocation whether working with film or digital. When the entirety of an endeavour is replaced with computer code, there’s nowhere left to go. Visit LinkedIn and look at the people who’ve used AI to create an idealised image of themselves rather than seeking a photographer or plastic surgeon. I once Googled a very alive, very real athlete, and the first picture of him was an AI slop photo. In what way is any of this good?

After the invention of writing, the invention of photography, regardless of the seachange they brought about in the vocal poet and royal portrait industries, they still required human beings to do the work. I’m a big fan of cats, but in relying on human beings to do everything for them, their brains have gotten smaller. I want to make it clear that a smaller brain does not necessarily mean they have gotten dumber. Brain size and intelligence are only loosely correlated. What it does correlate to is an adaptation to their new meta, one in which certain traits have atrophied out of existence, resulting in a physical change to their bodies. You’re free to make an argument that cats have better lives under us than fending for themselves, but even the smartest cats are frequently outwitted by doors. At no point was a cat capable of creating art, there are no cats on the liner notes of Sketches of Spain, and their literary output is shit. This is really no big loss to cats, and their base functions, hunting, sleeping, sharpening their nails, scratching at the door to be let out and then not going out, are working fine. When humans mentally atrophy, things are lost. Literacy, numeracy and problem-solving abilities have declined in the first world. We are in the golden age of the public person, and he’s increasingly a fucking idiot. I don’t know, but I would blame the technology before anything else, and adding a machine to do our thinking for us doesn’t seem like it would help things.

You’re posting on the internet, though

Yup.

Your opponents (probably) aren’t that stupid

☝️ Some of them will be, maybe astonishingly so, but by and large they’re probably just wrong. People can be intelligent and wrong. Sam Harris is notionally intelligent, and look at his incorrect take on the war in Iran. Hermann Göring had an IQ of 138, and the more I learn about that guy, the more I don’t care for him. I personally blame the 2000s Daily Show for this phenomenon. George W. Bush was dumb, so everyone who supported him was dumb, and everyone who worked for him was dumb, and this became the philosophical bedrock of the PMC liberalism lost to Trump, twice. What is actually lacking, specific individuals aside, is that people on either side of the AI debate come from completely different educational backgrounds, resulting in massive gaps in their respective learning.

Humanities education has been maligned for decades, purported to afford no jobs, no money, and no bitches. Instead, you should study something worthwhile and important, like accounts receivable and really make a difference in the world. Historically, the academy was intended to provide a rounded, multifaceted scholarship, to produce a person more in line with the notion of a Renaissance man than a Qualified Statistician. Someone would receive tutelage in both chemistry and poetry, mathematics and music, to have these blind spots in your knowledge base was brutish and uncouth. Isaac Newton, who invented calculus before he was twenty-six and is quite possibly the smartest human being who’s ever lived, was as immersed in theology as he was in physics during his matriculation. Nowadays, social and cultural pressures would force him to pursue one or the other. This would be no great loss to theology as the man’s ideas were a little odd, but a tragedy for physics and, by extension, our fundamental understanding of the universe. It would be a great loss to Isaac Newton, as well. His understanding of the many facets that go into the human experience would be diminished. If he were born today, there would be a serious chance that Isaac Newton would just be an epic Reddit atheist.

During the 20th century, and especially after the war, tertiary education began to become more about finding a job. Industrial capital needed skilled professionals in massive quantities; the more specialised, the less likely to question anything, the better. People would enter the workforce, perhaps as a very good trends analyst, with little knowledge of anything else. Have you ever wondered why your manager seems like a moron but has consistently achieved a 13% increase in productivity every quarter for the past fifteen years? It’s because he has a Bachelor’s in 13% Productivity Increases, yet can’t point out Asia on a map. It’s not that you must be an expert on everything; I personally am an expert on nothing, it’s also not that the apex human is an 18th century dilettante. It’s that when you charge someone tens of thousands of dollars during their early twenties and tell them that the rest of their lives depend on picking the right career, a decent portion of them are going to ‘lock in,’ and drown out the background noise of other fields of study. This results in a min-maxed society of people who do not understand the perspective of the others they encounter, beyond not wanting to speak to one another, they simply can’t. Their values, ideologies, and understanding do not click, and do not compute.

Conversely, most people, like myself, who oppose the AI push can’t even begin to explain how one is even put together. I made an effort to learn how diffusion operates, and even then, it is a surface-level understanding of what happens when you push ‘Go.’ Do I understand the programming that goes into one? Fuck no. Should I? Probably yes. Does anyone without a computer science background understand them? No, not at all, and it’s increasingly necessary that they should. I don’t mean in the facile ‘learn to use AI, or you’ll be left behind’ sense, there’s nothing to learn about ‘using’ AI, you tell it what to do, and it sometimes does it. I mean that we have organised our entire society around networked computers, and people don’t know how that operates. People don’t, won’t, or can’t learn even the basics of the scaffolding of the modern world. They then engage in a debate about what AI can and cannot do, sometimes with people who are simply fans of tech, and sometimes with actual knowledge. They end up making recourse to unhelpful statements like things lacking ‘soul’ and get dismissed as wet-brained sops. Their perception of an ordinary advocating for AI is that he’s a sociopathic progress obsessive without a concern for human lives, and they present themselves as ignorant nimbys.

It’s good fun to vent and abuse hyperbole online. Frankly, I think hyperbolic rhetoric directed towards the tech sector is not only cool and good, but a requirement for the future of the species. But most people you encounter aren’t actually working for Palantir, and as such, should be treated more kindly. People with worse tech literacy than I have are, loudly, advocating and actively pushing for more integration of AI into our lives, our workplaces. To be wary of such a sudden, dramatic change isn’t to be a Luddite; it’s to retain control of one's senses. You aren’t stupid for advocating for caution when upending society, the same way you aren’t stupid for thinking that a new machine could bring some benefits to society. What matters is who has control of that machine, and all of us, pro or anti-AI, should be concerned about that.

Yes, it matters that the tech sector is evil

I think we went astray when we deemed it gauche to shoot the messenger, especially when he’s the same guy who sent the message. The pompous entitlement of the tech sector’s pronouncements has only grown more calcified in the past few years. We let them get away with posturing as Adrian Veidt throughout the 2010s, even though their greatest accomplishments were building a website, something hip 90s teens would do in an afternoon IT class. As such, once the Geocities oligarchs had enough money to pay smart people to start working on machine learning algorithms, billions didn’t bat an eye as they declared that this was the future, this will be good, and there’s nothing you can do to stop it.

Despite how atomised we are these days, the result of the growth of the tech sector is that there appears to be fewer websites that ever. The majority of human online interaction takes place in about one of five places, each of these places owned by incomprehensibly wealthy and unaccountably powerful people. By existing in these spaces, we afford them a sort of populist endorsement. “Well, if the tech sector is so bad, why are you on Facebook/Twitter/Reddit?" Because we have to be, I’ve heard stories of people being denied a second interview because they weren’t on LinkedIn. We’ve moved the entirety of the species online, and there’s no easy way to undo that. It doesn’t matter that most people opened a Facebook account before their frontal cortex had formed; they’re there now, and so are all their friends. They work 50 hours a week for piss-all money and are too exhausted to go out on their days off, so they chat to their friends on Messenger, or ratio celebrities on Twitter, or try to make a buck selling pictures of their butthole on OnlyFans.

But in offering this populist endorsement, we endorse the companies that run them, and the people that run these companies, and there is something seriously wrong with the people who run tech companies. You might say it’s something wrong with the billionaire class as a whole, which is true; they shouldn’t exist, but there’s a particular repulsiveness to tech leaders that everyone feels but can’t quite place their finger on. They keep saying that they either want to wipe out humanity or that AI might wipe out humanity, and they keep working on it anyway, as if they don’t care. They fell in behind fascism with such enthusiasm that they shouldn’t be welcome in any sort of polite conversation. The billionaire class has done this before, but I can’t think of an instance of an entire industry ever doing it. They keep telling people it’s going to cost them their jobs, and in many instances it already has, yet they, and AI’s supporters, act so affronted when they receive even the slightest amount of pushback that this might be wrong. Now, of late, the attempts to murder them have forced some of them to walk back a few of their statements, but if they had the gall to make them in the first place, which one would you consider the lie?

They act with such total disregard for humanity that it’s getting to the point where they appear to be an active antagonist to the species. A human computer virus, born of podcast brain and corporate culture and that incredibly overrated Blade Runner sequel, and one that we at some point need to find a vaccine against. The environmental impact of data centres has been well documented. Beyond the whole water-usage thing, they are driving up power bills for regular people, increasing carbon emissions, and heating the land around them. All this so we can, what? Exterminate the human race? Who’s going to be using the AI if the planet dies? How would it even run? What a wonderful industry, what leaders, what intelligent people, we should listen to them. If this technology really has the power to do even a fraction of what it is claimed to be capable of, it needs to be taken out of the hands of creatures with complete contempt for humanity. It needs to be owned by the people, and the people need to make decisions as to how, when and where it is used. You might call that socialism or whatever, but ideally, it would be taken by force without compensation. So it’s more like communism.

When you make real art, the planet doesn’t take a hit, people’s livelihoods aren’t upended, and I feel like I’m going insane even having to say this. People’s power bills don’t go up because of a portraiture exhibit. You aren’t suddenly unable to drink the water near your home because of someone’s latest EP. An industry that does this, without concern, and demands applause for doing so, is psychotic and can never be reformed. If it could be, they would have never done it in the first place. If there was a shred of good in the modern tech bro, it died around the invention of social media, and it is not coming back. Remove their money, remove whatever benefits AI may or may not have, remove however much you enjoy your online spaces and latest gadgets. If a nerd in a Gestapo uniform came up to you in the street, and said he was offering you a free machine that meant you’d never have a job again, would damage your ability to think, and would set the planet on fire, would you accept it? What if he forced you to accept it? Would you thank him? Would you consider him a good person, a herald of the next phase of humanity? Maybe you’d think he and his machine were evil.

I’m done (Preach)

AI art isn’t bad because it lacks some je ne sais quoi. It’s bad because it’s not art. It’s bad because it’s not very good for you. It’s bad because the people who are giving it to you are not acting in good faith. You aren’t going to be left behind by not using it, you’re probably going to be ten times smarter, more empathic, more alive than the person who gets in the Amazon pod to prove they’re not a Luddite. It’s trite to say ‘pick up a pencil’ but the fact is you either pick up the pencil or let yourself be consumed by an industry that hates you, and a machine that will always produce its own things for you, rather than you producing something yourself.

That said, the baseline technology behind AI could be great, told you I’d be fair. There are applications in medicine, prosthetics, conservation, which run the gamut from exciting to fucking cool. Yet, for whatever reason, the push seems to be that first we make everyone unemployed, we strip people of their occupations, vocations and passions, and then later for all that other shit. Corporations jumped for joy when they learned they could use AI to wipe out their workforce. Money was spewed into every AI company, and every company that said they were going to cut their staff and install AI saw their stock price rise. Imagine if it weren’t private property, imagine what we could do if the machine were used to actually serve humanity the way it should? “Why have you saved the good AI can do for when everyone’s stopped reading?” Why has the industry saved all the good AI can do for after they’ve killed us all? Also, the pope fucking hates it, so you’re a bunch of heathens.