From what I read on the internet, people assume AI generated art is a difficult question legaly speaking. Some literally assume artists complain only because there are out competed.
I disagree - I think that AI generative art is an easy case of copyright infrigement and an easy win for a bunch of good lawyers.
That's because you can't find an artist for a generated picture other than the ones in the training set. If you can't find a new artist, then the picture belongs to the old ones, so to speak. I really dont see what's difficult with that case. I think the internet assume a bit to quickly it's a difficult question and a grey area when maybe it just isnt.
It's noteworthy that Adobe did things differently than the others and the way they did things goes in the direction im describing here. Maybe it's just confirmation bias.
> I disagree - I think that AI generative art is an easy case of copyright infrigement and an easy win for a bunch of good lawyers.
> That’s because you can’t find an artist for a generated picture other than the ones in the training set.
First, that’s clearly not true when you are using ControlNet with the input being human generated, or even img2img with a human generated image, but second and more importantly…
> If you can’t find a new artist, then the picture belongs to the old ones, so to speak.
That’s not how copyright law works. The clearest example (not particularly germane to the computer generation case, but clearly illustrative of the fact that “can’t find another artist” is far from dispositive) is Fair Use noncommercial timeshifting of an existing work: it is extremely clear there is no artist but that of the original work, and yet it is not copyright infringement.
> I really dont see what’s difficult with that case.
You’ve basically invented a rule of thumb out of thin air, and observed that it would not be a difficult case if your rule of thumb was how copyright law works.
Your observation seems correct to that extent, the problem is that it has nothing to do with copyright law.
> I think the internet assume a bit to quickly it’s a difficult question and a grey area when maybe it just isn’t.
IP law experts have said that the Fair Use argument is hard to resolve.
Assuming the lawsuits currently ongoing aren’t settled, we’ll know when they are resolved what the answer is.
It’s not as simple as that though because the algorithm does learn by itself and mostly just uses the training data to score itself against, it doesn’t directly copy it as some people seem to think. It can end up learning to copy things if it sees them enough times though
“you can't find an artist for a generated picture other than the ones in the training set. If you can't find a new artist, then the picture belongs to the old ones, so to speak”
I don’t think that’s valid on its own as a way to completely discount considering how directly it’s using the data. As an extreme example, what if I averaged all the colours in the training data together and used the resulting colour as the seed for some randomly generated fractal or something? You could apply the same arguments - there is no artist except the original ones in the training set - and yet I don’t think any reasonable person would say that the result obviously belongs to every single copyright owner from the training set
Normally - outside the specific context of AI generated art -, there is not a relation "work¹ → past author" , but "work → large amount of past experience". (¹"work": in the sense of product, output etc.)
If the generative AI is badly programmed, it will copy the style of Smith. If properly programmed, it will "take into account" the style of Smith. There is a difference between learning and copying. Your tool can copy - if you do it properly, it can learn.
All artists work in a way "post consideration of a finite number of past artists in their training set".
But this person’s dog isn’t in the training set, so why should some artist be credited for a picture they never drew? Not a single person has drawn his dog before, now there is a drawing of his dog, and you want to credit someone who had no input to the creative process here?
"Input into the creative process" is surely broader than simply "painted the portrait". Artists most certainly never consented to have their works used as training data. To this extent, they might be justifiably pissed off.
Artists and designers have furthered their careers (and gained notoriety) by 'ripping off' others since the dawn of time. This used to require technical artistic ability; now less so. The barrier to entry is.... not necessarily lower now, but different.
That’s not (either positively or negatively) how copyright law works. You can not have a new artist and still not be infringement, and you can have a new artist (as is the case for derivative works by someone other than the original artist) and still be infringement if neither licensed nor Fair Use.
> That's because you can't find an artist for a generated picture other than the ones in the training set. If you can't find a new artist, then the picture belongs to the old ones, so to speak.
It doesn't belong to the "old ones", it is at best a derivative work. And even writing a prompt, as trivial as it might seem, makes you an artist. There are modern artists exposing a random shit as art, and you may or may not like it, but they are legally artists, and it is their work.
The question is about fair use. That is, are you allowed to use pictures in the dataset without permission. It is a tricky question. On one extreme, you won't be able to do anything withing infringing some kind of copyright. Used the same color as I did? I will sue you. On the other extreme, you essentially abolish intellectual property. Copying another artist style in your own work is usually fair use, and that's essentially what generative AI do, so I guess that's how it will go, but it will most likely depends on how judges and legislators see the thing, and different countries probably will have different ideas.
I don't believe it's a tricky question at all: "Did you train your model on an artist's work that has copyright protection without permission?" is the very simple, straightforward question. The fact of its being an AI model is irrelevant.
Did you train _yourself_ on an another artist's work that has copyright protection without their permission? Yes? That's ok, because copyright law doesn't care.
If you have an argument, spell it out. Machine learning is not human learning. They are not the same. Arguing "because human beings learn" does not lead to the conclusion "therefore training models on an artist's work and thereby producing work that looks like theirs is okay". Or, if that is your argument, you have to do more work than just gesturing vaguely at the word "learning".
>That's because you can't find an artist for a generated picture other than the ones in the training set. If you can't find a new artist, then the picture belongs to the old ones, so to speak
We have some countries where it is explicitly legal to train AI models on copyrighted data without consent, and precedent in the US that makes this a plausible outcome there as well.
Could you explain what portion of copyright law you believe would cover this argument? I'm not a lawyer, but have a passing familiarity with US copyright law, and in it, at least, I do not know of anything that would support the idea you're proposing here. How would you even assign copyright to the "old" artists? How are you going to determine what percentage of any given generation was influenced by artists X, Y, Z?
> AI generative art is an easy case of copyright infrigement...
Agreed. An AI model trained on an artist's work without permission is IP infringement and this should be widely understood. Unfortunately, because the technology is new people do not understand this. When Photoshop was new, there was a similar misunderstanding. People could take an artist's work, run it through Photoshop, and then not compensate the artist. It took some time for that to sort out.
I agree. This is a clear-cut case of copyright infringement, as is all art. After all, people painting images have only seen paintings other people painted.
There's copying and there's being inspired by. We don't know where AI generators fall yet, legally. The GP's argument of "they've only been trained on images, so everything is infringement" is a fallacy. Humans have been trained on images too.
The fallacy is thinking that macine learning is like human learning. It is not the same at all, and yet people continuously conflate the two whenever discussions like this come up. "Because humans learn, there is no problem with AIs being trained on an illustrator's entire corpus so that it outputs works convincing enough to be a forgery" is the argument you are making and it's a bad one.
Why does it matter how well it works? If a human produces a forgery, does that constitute copyright infringement? If it's inspired but sufficiently changed, is that a new work?
There's already a line, what does it matter who produced the work? Just judge everyone using the same criteria.
"convincing enough to be a forgery" was the deliberate choice of phrase not "forgery". An illustrator who spent many unpaid hours to hone their craft should get to enjoy the fruits of their hard work. A kid with a GPU and a model who trains it on that artist's corpus is stealing, not "learning". Every illustration from that model is theft.
Even if you don't find the moral argument convincing (and you should), legally protecting creators will encourage creativity, which is the whole point of copyright, while allowing anyone with a laptop to coopt their work will discourage it.
If your argument is that everything produced by an AI is theft, I don't think we'll agree. I don't even know how we'd begin to tell who each Midjourney image stole from.
The only problem to that, and a big one, is that there’s no way to trace back to the image in the dataset from a final output of AI.
It’s a static mapping, surely it should be possible, you’d think, but NN frameworks aren’t designed that way. That is blocking it from happening(and also allowing “AI is just learning, human is same” fallacy)
I disagree - I think that AI generative art is an easy case of copyright infrigement and an easy win for a bunch of good lawyers.
That's because you can't find an artist for a generated picture other than the ones in the training set. If you can't find a new artist, then the picture belongs to the old ones, so to speak. I really dont see what's difficult with that case. I think the internet assume a bit to quickly it's a difficult question and a grey area when maybe it just isnt.
It's noteworthy that Adobe did things differently than the others and the way they did things goes in the direction im describing here. Maybe it's just confirmation bias.