GenAI Art Is the Least Imaginative Use of AI Imaginable

You aren't wrong, especially about motivations. The "current consumer focus" has had some weird results, too - particularly in the image/video generation space, it's the consumers, or rather the OSS community, that actually did 90% of the work of turning AI research output into product prototypes - which many a startup now just pick, polish and scale up. Not to dismiss that work, either, but that's mostly the old-school SaaS building work; the hot AI part is already there on Github, packaged into ComfyUI nodes for people to play with.

Full video is a slightly different story because of compute demands, but image generation is basically the infamous Dropbox comment situation - any fancy image generation task you see offered as a service, a moderately tech-savvy person could replicate for personal use with ComfyUI over a weekend. That applies to interactive portraits too - the nodes are but a click away, models run fast enough for local use on an average PC.

I'd like to challenge you on your conclusion, though:

> These tools are purely intended for superficial people who are not interested in creating things, they are only interested in outputs and end results.

But this is exactly how progress is made. Quoting Alfred North Whitehead[0]: "Civilization advances by extending the number of important operations which we can perform without thinking of them." Where for one craftsman, the process is the most important part of their creative work, another craftsman only cares about the output of the former, because it's an input to their creative work. That doesn't make the latter person "superficial", they're just one level up the dependency graph.

--

[0] - https://en.wikipedia.org/wiki/Alfred_North_Whitehead

Alternatively current AIs don't compose well with each other and don't let themselves be decomposed easily. This makes control and user interaction difficult to establish which leads to monolithic one-click input->output products by default.

Things that can be stated simply ("replace the banjo with a saxophone", "place the cat to the left of the table") can't really be tackled by composition (comfyui workflows of hell) and if you want something that generalizes, you need to train models specially designed with that composition scheme in mind (layout diffusion for instance).

It's not the AI that makes the art, but the person who tells the AI what to do. A painter friend of mine has gotten into generative AI lately, and the sometimes-pretty, sometimes-intimidating pictures she comes up with sure look like an artistic expression to me. She has a vision of something she wants to see, and she uses tools to bring that vision into reality, just like she does with paint; the new tools just give her new options.

As far as I understand, art is typically understood to have two components: expression and impression. The creation of something, anything, as a form of self-expression, is considered art. Additionally, though, art is in eyes of the beholder. If a feeling is inspired in someone by the observation of a thing, act, etc, then it becomes art to them, in that moment. Notably, though, it doesn't necessarily make that thing art to anyone else, unless it is curated at which point that curation is a form of expression.

These definitions are distinct from the requirements for copyright.

>Self expression is not only inherent in art, it’s essential.

That idea is only about 200 years old and strongly Western. The vast majority of cultures saw no difference between "art" and craft, regarding pictures and sculptures as simply decorative objects, of the same basic cultural standing as pottery or textiles. Our understanding of "art" was fundamentally transformed by the development of photography and mass-manufacturing.

Anyone with a glib answer to questions about the cultural meaning of AI-generated art really needs to familiarise themselves with the thinking of William Morris and Yanagi Sōetsu, because this isn't the first time we've had this argument.

> It’s possible for someone to make art from GenAI outputs, but that requires taking artistic liberties post generation.

Plenty of people use others' expressions as a form of art, (sharing song lyrics, singing others' songs in the shower, cat memes, and so on).

It is all the genius really of Duchamp's Fountain from 1917.

At this point the conversation is just boring.

Sure it isn't art but I guess it will be a substitute for art for many things: do people care that much if their workout music is actual art or artificial not-art art-like stuff?

> I would really like an objective explanation of why this attitude "that the whole point of art is the process of making it" is anything but pretentiousness in it's purest form.

i think that's a problem - i am no philosopher or art historian, but i truly can't think of "objective" truths of any kind when i think about art, the process of making art, or the process of engaging with art. it's nice to make art. i play around with synthesizers and it... makes me feel more like a human being, i guess? when i record music or do amateur photography i've actualized myself somehow. something that wouldn't have existed otherwise now exists because of my actions, and that's enjoyable to experience. i imagine one could argue "you still created something that wouldn't have existed when you prompt an AI!" but it just doesn't feel the same, and i struggle to explain why

> And even if one supposes that a key part of the value is self-exploration

i think a lot of value is the context of the art itself - picasso going insane, the protests leading up to the tank man photo, aphex twin's cheetah, jesse krimes being in prison, whatever. maybe this is just me, but when i go to museums the extra information in the labels or audio descriptions is where i find most of the joy and fascination that makes the artworks meaningful to me. AI art feels contextless

> Surely the actual artifact itself has some value?

genuinely - have you encountered AI-generated art that you find valuable? that you tell other people about or have kept a copy of for yourself somehow?

> I would really like an objective explanation of why this attitude "that the whole point of art is the process of making it" is anything but pretentiousness in it's purest form.

It's not necessarily pretentiousness (though it's definitely pretentiousness adjacent), but a view from the point-of-view of an artist. I say it's adjacent to pretentiousness because I think implicit in the entire article is the assumption that creation is superior to consumption (and also the assumption that art created just to pay the bills is inferior to art created for its own sake[1]).

1: https://en.wikipedia.org/wiki/Potboiler

“So what," the Chelgrian asked, "is the point of me or anybody else writing a symphony, or anything else?"

The avatar raised its brows in surprise. "Well, for one thing, you do it, it's you who gets the feeling of achievement."

"Ignoring the subjective. What would be the point for those listening to it?"

"They'd know it was one of their own species, not a Mind, who created it."

"Ignoring that, too; suppose they weren't told it was by an AI, or didn't care."

"If they hadn't been told then the comparison isn't complete; information is being concealed. If they don't care, then they're unlike any group of humans I've ever encountered."

"But if you can—"

"Ziller, are you concerned that Minds—AIs, if you like—can create, or even just appear to create, original works of art?"

"Frankly, when they're the sort of original works of art that I create, yes."

"Ziller, it doesn't matter. You have to think like a mountain climber."

"Oh, do I?"

"Yes. Some people take days, sweat buckets, endure pain and cold and risk injury and—in some cases—permanent death to achieve the summit of a mountain only to discover there a party of their peers freshly arrived by aircraft and enjoying a light picnic."

"If I was one of those climbers I'd be pretty damned annoyed."

"Well, it is considered rather impolite to land an aircraft on a summit which people are at that moment struggling up to the hard way, but it can and does happen. Good manners indicate that the picnic ought to be shared and that those who arrived by aircraft express awe and respect for the accomplishment of the climbers.

"The point, of course, is that the people who spent days and sweated buckets could also have taken an aircraft to the summit if all they'd wanted was to absorb the view. It is the struggle that they crave. The sense of achievement is produced by the route to and from the peak, not by the peak itself. It is just the fold between the pages." The avatar hesitated. It put its head a little to one side and narrowed its eyes. "How far do I have to take this analogy, Cr. Ziller?”

- Iain M. Banks, Look to Windward

Most artists create art that nobody cares about. This is just reality. You can make music, and likely nobody will listen to it. You can write a book, and probably nobody will read it. You have to create for yourself and if you’re extremely lucky someone else will care, but probably not.

AI art is a cute trick, but it makes this point even more true. Why would you care about a picture I drew when you could generate 100 of them in minutes and way better as well? The fact that you can put a coin in the vending machine and get a picture or song or novel completely devalues the end product.

What remains is the process of creativity, of choosing a medium and exploring the boundaries of what it can do. I saw a Picasso exhibit that showed all the studies and sketches he’d done in preparation for a piece. That process IS the creation of the art. The final picture isn’t even that interesting.

If someone feels fulfilled generating images, good for them, I guess. To me it seems like a cheap fix and a trick that we will all eventually grow tired of.

The point of art for the artist is expression. The point for the audience is to gain something through shared experience with the artist.

AI outputs are not created by a conscious agent experiencing affect, and its audience shares an experience with no one. It's not compelling, and isn't art.

  To me, this is a rationalization for disliking AI because it displaces humans to some degree or another. I think that is a fair reaction, but the rationale is nonsense.

AI images and users gets filtered out everywhere, closing into an edgy echo chamber groups, and rarely, if ever, for such "it displaces humans" philosophical arguments. AI generated data just evokes negativity, are low quality, and are unsalvageable, and there are little to no prospects of that improving through cooperation because AI-leaning people have ran around setting so many bridges on fire, so its potentials are all but ignored.

If anything, I'd expect the resentment will subside over the next decade, as generation techniques mature and barrier to entry for amateurs exponentially rise. I think we are already seeing signs of such guild moat forming in code generation LLMs.

You are exactly right, the artists are deploying all their rhetoric skills they learned in university to try to avoid becoming the next horse drawn carriage drivers.

Underappreciated comment right there. The idea about selection vs creation was on my mind, but I wasn't able to put the idea into a solid form as you did there. Thanks.

Kinda understandable, but as a consumer, I already dislike that a lot.

In the beginning, AI 'art' was novel and interesting in a way. But now I've already seen so much generic illustrations, and in 95% they don't add any value to a blog post. I'm even less likely to click on links with a generated thumbnail these days. It creates an expectation of SEO spam or low quality and I have to wonder if the author also used AI to write the text. The generated images seem so soulless to me.

This is not creating art, but illustration. Art is where the meaning is inherent to the work, whereas illustration is when it is derived from something else that it accompanies (e.g. a text). There’s obviously a spectrum here, but GenAI blogpost decoration are pretty much the terminal point on the illustration part of that scale.

The world is topsy turvy. I can't even tell satire anymore or whether people are being sarcastic and tongue-in-cheek. This comment reads like sarcasm to me, overdoing on the things currently crappy about online "content creation":

0. The word "content" itself and self-applying the "content creator" label. No writer, artist, or musician I know thinks of themself unironically as a creator of content. How bland! Like a manufacturer! I produce units of content. My buddy who plays the saxophone and does a little composing would barf if you told him he was an audio content creator.

1. Feeding the algorithm. Do people really think about what "the algorithm" wants when they are being creative? Do they alter what they make to suit the algorithm? Why care what an unfeeling search engine wants if you're ultimately writing for the human soul?

2. Engagement as a goal. Is this really what you're after when you write? Not to inform, educate, please, inspire, upset, challenge, upend, or some other meaty goal? You just want some kind of generic "engagement"? This is the kind of thing a Marketing SVP wants, not a writer!

Sorry, I don't mean to attack. I'm trying to avoid being cynical, but the comment really struck me. I can't believe someone would write something like this about their creative passion, to the point where I question that maybe it was amazing sarcasm.

What you've described is the exact mechanism through which generative AI will actually kill search on the open web.

It's less likely that search will be threatened by the alternative of Q&A with a chatbot as much as by slop overrunning valuable content. More and more of the discoverable content will be machine-generated filler produced for exactly the reason you describe, it's rewarded by rankings.

People won't even bother writing on their own, let alone producing images to go with the writing. It'll be like the herculean task of just finding a damn recipe online without reading someone's life story first, but 1000x as bad. It's already here I guess. We've seen the baby peacocks in the coalmine.

Once I was an editorial illustrator, and the mantra always was "concept is king". Stick figures are kosher so long as they communicate something.

Impressionists emerged following the invention of the camera, as they had no more incentive to pursue realism. I think in the age of AI slop any form of crude image making is better so long as it's human. Scribble something out with crayons. Shape an animal out of a slice of bread, post a pic. It's all legit now, technical art abilities mean nothing. To impress you actually have to have a good idea.

Isn't it amazing when a good chunk of monetary incentive in the present day is directed towards the terrifying concept of "engagement"?

Use free stock photos instead. Your content will look less spammy.

Would you consider the images you create to be art?

Although the chatbot in chrome devtools is really useful. I selected a DOM node and ask "why isn't this visible" and it queried the DOM, css and figured out the height was 0 and explained exactly what lead to it being 0.

Google Photos just added this completely subpar feature for how I have to search my photos in the app. -.-

I bought a book a few years ago called The Artist in the Machine by Arthur Miller[0] and it covers a lot of early AI creativity. I wish they shelled out for more color pages, because I agree things like DeepDream and GAN art are very trippy and would have made it a great coffee table book. It's wild that it took just two or three more years for diffusion models to hit the scene and take the world by storm.

[0] https://direct.mit.edu/books/book/4547/The-Artist-in-the-Mac...

Deep dream! The eyes

What about NFTs as registrations for domain names a la ens.domains?

Absolutely

None of those links are art. They’re stunts at best. I know, I know someone will disagree, but they’re wrong.

The difference between cave paintings, brush paintings, and even photoshop versus AI generated works is that a human had to visualize mentally and then translate their internal visualization into the medium without words. AI generated does not require that.

Artists like Jackson Pollock or more modern instances of digital artists lie somewhere between the two—creating with little intention and little control.

Inventions like the dishwasher for domestic use never really led to meaningful work reduction for the people (women) doing this kind of work.

That is not the tradeoff being made. When you go from a home without a dishwasher to one with a dishwasher, you don't work more or less hours as an engineer at JPL. You work the same hours. Likewise for someone who does their housecleaning by hand or one that hires housecleaners, the latter doesn't start working 50 hour shifts at the office in response.

Then the question is what are people doing with the extra time? We'd all like to imagine we are an artist or poet or musician deep own if we had the time, but are we really? Or would most people take an extra ten minutes and use it not to pick up a violin or to work a garden, but to scroll through tick tock perhaps?

And you have to at least be aware to the fact that there is a lot of money to be made trying to get you to scroll tick tock instead of playing a violin not looking at advertisement, and very few interests out there pushing back as a result. So what gains we might make in efficiency today are liable to be gobbled up by some sort of profit making addictive product. And personally speaking, sometimes these low effort tasks like dusting the home or cleaning the dishes or mowing the lawn are great ways to disconnect active process into an automatic sort of action, to mediate while you are doing the thing, and think inwardly about some other idea you might have. In some ways washing the dishes is a forced chunk of time you actually have to think your thoughts. And scrolling an algorithmic feed in its stead is very much not giving yourself a chance to think your own thoughts about your own situation; you've in fact offloaded your thought process to the highest bidder to think for you. And that seems quite dangerous in society.