I first dabbled with AI image generation back in 2022 and sprinkled a few such images throughout my worldbuilding project. It was easy to look past all of the flaws with the idea that it was nothing more than a novelty. And I never cared nearly enough about my worldbuilding to pay anyone for artwork of it.
Now that I look back at it, those images are obvious slop, which I've grown to dislike as much as the next person. But recent comments I've seen here and on other sites have made me wonder if my brain has rotted in the same manner that makes some boomers fall for AI slop. There will be videos where the use of AI is not very noticeable to me, but not with deceptive intent. Maybe an illustration to get the point across or a subtle two-second animation. Commenters will very passionately point it out. To be honest, I don't see the creator either paying for the equivalent human work or drawing anything better themselves.
Does it really just look that bad? Is it an issue with what AI and the companies that sponsor it stand for? Theft of real artists' work? Does it change at all if the images were generated locally with the creator's own hardware and resources? What about upscaling images, like I do with old wallpapers so that they look better on new monitors?
I assume what I've just said will attract downvotes, but that was my thought process and I do want to understand where other people draw the line and for what reasons. Should we limit it to quick-and-dirty illustrations, pure novelty, upscaling existing images, a model that only incorporates work if the artist consents, or something else?
never. ai "art" is theft.
Let's not just exchange blunt claims, but reason a little.
Copyright critics have long made the somewhat compelling argument that copying isn't stealing because the original digital item does not become scarcer in the process. So how can AI taking artists' work be considered theft if it, too, just uses copies of the original work and maybe transforms them into a new work (which would, under U.S. law, fall under "fair use")?
We might argue that, well, fair use does not apply because most AI companies try to monetise the models derived from other people's work.
Which leads me to the question: would you find visual genAI more acceptable if it weren't commercial?
As bad as copyright laws are, they are made even worse when they are applied to everyday people via life-altering fines, and often ignored completely when applied to giant corporations.
I get and share the criticism of double standards in the application of the law and the other ugly sides of corporate AI. The question I'm asking here is: if unshackled from its corporate contexts (i.e. proprietary models run for-profit in centralised data centres): is genAI still objectionable? Is the tech unethical as a wholez or is it only problematic because for now, it is mostly a tool of the oligarchs?
We wouldn't even call it AI anything then. The whole framing as it is now is to undermine creative work. It's hard to imagine how it would even remotely resemble what it is now, a cold reading con.
Because AI companies are making a huge profit (and wrecking the environment) on stolen art.
Thereβs a huge difference between someone that torrent 1T of books over his/her life and a company that torrent every books in history to make a machine that will try to destroy art as we know it for the profit of the 1% few.
Alright, I get that. It's pretty much what I wrote up there in the second half of my posting.
I'd still be curious as to your answer to my question there:
Personally probably still see it as slop, because I think that genAI are just a by product of the inherent spyware they are
For sure. I couldn't tell you where inspired, but new, work ends and theft begins, or how model training would be funded without commercial incentive. But I would be more comfortable knowing that companies have not ripped potential profits straight from every artist the model had been trained on.
And I like this kind of discussion. What's bothered me before have been the jabs at the mere presence of AI without deeper discussion as to what qualifies as theft. I haven't found myself buying art even before such AI models. I wouldn't buy or sell images that I know are AI generated. And I pay the electric bill for locally-generated ones as if I were doing any other novelty activity like gaming on my PC. I'm curious, what would you think of local models that can be acquired for free?
I'm probably the most anti-AI person I know, but I agree discourse around how "AI is theft" is a bit shallow.
However.
Copyright is often erroneously conflated with plagiarism. While the two do sometimes coincide, they're very different concerns.
I, myself, believe copyright is so broken we'd be better off throwing it away. (The only thing I believe I'd miss about copyright if I woke up tomorrow and it didn't exist would be copyleft.) But I do deeply believe in a right to attribution.
I don't think AI is theft. I think it's plagiarism.
And I do believe that listing the names of all those whose works were involved in training data for a model would still be a great disservice to the artists. Meanwhile, asking an LLM or image generating model which training data was involved in generating the particular output it produced is similarly futile to asking which rep at the gym allowed a stage strongman to lift a car.
And if someone objected that giving what I would consider "sufficient credit" to artists/authors/whoever would make AI models completely infeasible, then my response would be "that's exactly my point."
Finally, one more point I want to make is that if AI didn't make billionaires a huge amount of money, the legal system would have put a stop to the mass scraping of training data and made a very visible example of whoever undertook to do so first long ago. (coughAaron Swartzcough) As terrible as it is having to deal with the shitty IP laws we have, the greater injustice is that the laws only apply when billionaires want them to.