[dylan604]

i'm fine if the call it artist impression without calling it AI generated. i don't care if the artist's impression was created in watercolors, oil, charcoal, or AI. as long as it is identified as not an actual image is the main concern

[jacobgkau]

The problem is that it's an AI program's impression, not an artist's impression.

Edit: If you can't get past the fact that AI image generators are not equivalent as an artistic medium to different types of paint, think of it this way: if I describe something to an artist and have them paint it, is the resulting image my impression because I prompted the artist, or is it the artist's impression of what I described? Clearly it's the artist's impression. So just because an "AI artist" prompts an AI app, that wouldn't make it the prompter's impression; it's still the AI app's impression. And an AI app's not an "artist" itself by virtue of being a computer program and not a human (as you yourself admit by attempting to liken it to a tool such as paint).

[dylan604]

> The problem is that it's an AI program's impression, not an artist's impression.

I just made a similar reply, but I disagree with this. The artist iterates with their prompts to the AI tool to get what they wanted. So when they stopped tweaking the prompt, they were satisfied with the result to be their impression

[aithrowawaycomm]

"AI art generators enable the creation of ignorant and lazy illustrations by outsourcing understanding to an idiot robot."

"Yes, but is it not the intent of the artist to be ignorant and lazy?"

It is possible to repeatedly iterate AI art gen and get what you want, but that's not what happened here. And even so, it's not at all the same thing as drawing a picture: "iterating on what you want" is equivalent to curating art, not creating it. In the US you can copyright curation and that extends to curation of AI art - the US Copyright Office correctly said that tweaking prompts is the same thing as tweaking a Google Images search string for online image curation. But you can't copyright the actual AI-gen pictures, they are automatically public domain (unless they infringe someone else's copyright).

[dragonwriter]

> But you can't copyright the actual AI-gen pictures,

The US Copyright Office has, in fact, granted at least one copyright registration for an image created with the use of AI generation.

https://news.artnet.com/art-world/invoke-snags-first-ai-imag...

[aithrowawaycomm]

I am specifically talking about DALL-E or Stable Diffusion, your link describes something very different. The point was the "Google Images" analogy, which applies to 99.999% of AI art but this is an exception.

[dragonwriter]

> I am specifically talking about DALL-E or Stable Diffusion, your link describes something very different.

No, it doesn't. It describes artwork done on Invokeai, one of the popular hosted web frontends for Stable Diffusion (and some similar models), with a process very much like what many AI art hobbyists use (whether with hosted frontends or locally-hostable ones like A1111, Forge, ComfyUI, or Fooocus.)

[dylan604]

> But you can't copyright the actual AI-gen pictures

Is that actually true? What I’ve read is that AI can’t be credited with the copyright.

[aithrowawaycomm]

No, in general you cannot copyright them:

https://www.reuters.com/legal/ai-created-images-lose-us-copy...

https://www.reuters.com/legal/litigation/us-copyright-office...

https://www.reuters.com/world/us/us-appeals-court-rejects-co...

See the other reply for a half-counterexample, but the major difference is the specific software is more like generative PhotoShop, and the final image involved a lot of manual human work. Simply tweaking a prompt is not enough - again you can get copyright for curation, just not the images."

Of course AI can't be credited with copyright - neither can a random-character generator, even if it monkeys its way into a masterpiece. You need legal standing to sue or be sued in order to hold copyright.

[rendaw]

Isn't it also possible that it wasn't an artist at all, instead someone whose job was never illustrating scientific articles (like a manager or a random intern), who put the text in the prompt and went "that looks pretty sciency, good enough" and the person responsible for the publication went "great, we just got a sciency image and saved $XX!"

Yeah the intern is now an "artist" but I think lumping them together is muddling the discussion.

[setr]

When I dictate to a PM how they should present my stuff in their useless PowerPoints, that result could be called the PM’s impression. More specifically, their impression of my impression of the subject.

But after I’ve iterated with them N times, fixing their various misunderstandings of my impression, eventually the PM’s impression impact approaches zero, and it’s clearly representative solely of my impression.

Sufficient iterations and prompts against the model, and you can’t really say it’s the AI’s fault the result is erroneous. The first pass, sure, but not the 100th.

Except I guess in the aspects where you can’t fix the PM’s/AI’s understanding. Then you could say their impression isn’t fully removable

[johnnyanmac]

And why are we assuming that someone lazy enough to use AI isn't doing more than 1 prompt to make something and going with it? They've lost all benefit of the doubt in my eyes.

[setr]

I’m not making that assumption. I’m suggesting that the difference between AI impression and Artist Impression is number of iterations — the solution to this disagreement.

Both terms are acceptable to use; this argument is fundamentally about how lazy you should assume these users to be. So argue about that directly.

[LtWorf]

Except, despite all of your contempt project managers are people who can learn. LLMs are trained and can't learn anything after that. They have a very very short sliding window of context that will start to be dropped when you add more information.

So your example makes no sense.

[setr]

Why does it matter whether they can learn? If I let them run off after a single pass, then yes their understanding of my understanding is relevant. After the Nth review, it’s not. The latter is ideal, else you have the game of telephone

The question is whether their understanding still contributes to the end product, not who does the mechanical action of entering data or drawing images.

[LtWorf]

> Why does it matter whether they can learn?

Because nothing of consequence can be done in a single pass.

[setr]

…what? If I tell the PM the info needed, and he sufficiently produces the PPT without error, then it’s done? Why do I need further back n’ forths to make it consequential?

The number of passes/reviews is directly tied to error rate. If the PM/AI is able to produce my impression at 100% success, then great, there’s no further work to do.

The only thing learning matters for is with sufficient learning, they might reach a state where they no longer need to be reviewed, because they’ve sufficiently learned to not inject their own interpretation of things into it. They are now a straightforward extension of my own being, and have generated their documents as I would have done (had I the requisite mechanical time/skill/interest for producing whatever is in question).

[soerxpso]

The issue is the artist in this particular case didn't really provide their impression. When I hear that something is an "artist's impression" I expect it to mean that a reasonable thinking person read about the thing in at least more detail than I have (since all I've done is skim an article) and then created an image that conveys the information visually. If all the artist did is tell an AI to render "a piece of plastic dissolving in the ocean" it's not conveying any information that's not in the headline, and it's not very pretty either.

If the point was purely aesthetics, maybe they could at least use Midjourney or that new Ghibli style transfer thing instead of what looks about as good as old Dall-E? It's ugly. If the point was to convey information, they could have done that better with a 30-second pencil sketch, which would have also taken less effort than the AI-generated thing.

[trompetenaccoun]

You're fine with being mislead a little?

I'm not a materials scientist so I can't comment on this specific topic but based on my experience with pop science reporting errors and misinformation often come in multiples. The author has a "Bachelor of Arts in Professional Writing". RIKEN's press release is already written for a general audience and in English, so there isn't a good reason not to read the original source instead.

[dylan604]

yes, because it's not really a misleading title as it is still the artist's impression. If the artist didn't like it, they would just keep modifying the prompt until they were satisfied. So it is the artist's impression.

where are we disagreeing?

[trompetenaccoun]

Users of popular apps are not in control, the people tweaking the models are. GenAI is not an artist.