Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Database of artists used to train Midjourney AI garners criticism

artnews.com

Post reply on HN
Print view View original

61–70 of 144 posts

Re: Database of artists used to train Midjourney AI garners criticism

#61

Post by jncfhnb »

bjt wrote: post #33
tbrownaw wrote: post #30

Earlier quoted context omitted.

> status quo in the United States is that AI-generated images are not currently eligible for copyright. Aren't they? I thought it was just that the copyright holder has to be a recognized legal entity (so, the copyright would have to belong to the human operator or their employer, not to the ai model itself).

From the article: > Last September, the US Copyright Review Board decided that an image generated using Midjourney’s software could not be copyright due to how it was produced.

A grossly misleading oversimplification.

Re: Database of artists used to train Midjourney AI garners criticism

#62

Post by tavavex »

johnnyanmac wrote: post #38
skydhash wrote: post #14

Earlier quoted context omitted.

Lockpicking tools are widely available, and designs of locks can be found on the web if someone search hard enough. But entering another's home is illegal. It's why businesses pay for licenses even if cracked softwares exists. I hope that artists win these and make using or creating an illegal model an high-risk activity, not worth it for any commercial activity.

I don't even care about it being illegal. As said above, US can completely ban such tech and China will keep on trucking along. I just want artists affected to be paid, and actually paid. Not "paid" the way spotify artists are. if you're making a billion off of 100 artists' work, you better be making each affected artist a millionaire in royalties.

I'm a bit confused - you draw parallels between artists whose work gets used in a training dataset and musicians who sign contracts with Spotify, but the comparison is very strained. The financial link with Spotify is obvious - the company pays some minuscule amount of money per use to directly use a product of an artist, in part for their own profit.

With AI, the situation is far from being that clear-cut. For one, the majority of datasets aren't comprised of capital-A Art with specific authors and attributions - a ton of it is just random information from around the internet, or just downright junk data. A photo I posted for free could conceivably end up in that dataset. Then, when a model is trained on that dataset and that model is used to make other outputs, are those outputs actually derivative with a direct link to some origin? If I generate a landscape using an AI, to whom exactly would I even owe money? Equally to all contributors in a dataset?

Re: Database of artists used to train Midjourney AI garners criticism

#63

Post by Gigachad »

skydhash wrote: post #14
srackey wrote: post #4

I get why artists are trying to stop them, but this battle has already been lost. These tools have been out too long, open source models proliferate freely, and jurisdictions that don’t care about IP law will continue publishing these models. By the time it works it’s way through the courts the situation will be even worse.

Lockpicking tools are widely available, and designs of locks can be found on the web if someone search hard enough. But entering another's home is illegal. It's why businesses pay for licenses even if cracked softwares exists. I hope that artists win these and make using or creating an illegal model an high-risk activity, not worth it for any commercial activity.

The difference is that AI generation tools provide a lot of potential economic value, while breaking in to someones house doesn't. There will be a lot of pressure for governments to allow this to go on since it puts them at advantage over other countries.

Re: Database of artists used to train Midjourney AI garners criticism

#64

Post by teaearlgraycold »

semiquaver wrote: post #52
thurn wrote: post #3

One positive aspect of the status quo in the United States is that AI-generated images are not currently eligible for copyright. I think this is a great direction to go in, I highly doubt Wizards of the Coast or whoever is going to want their premium products to lose copyright protections, so they'll need to keep paying artists. I'd love for us to lean into this -- you can make all the AI art you want, but it automat…

> AI-generated images are not currently eligible for copyright It’s a bit more nuanced than that. Here is the relevant policy statement, which notes that some AI-assisted works are potentially eligible for registration and have indeed been registered, while works that are primarily the product of an AI are not. https://www.federalregister.gov/documents/2023/03/16/2023-05...

Makes sense. Photoshop has had content-aware-fill for over a decade. That counts as AI as much as any diffuser does. I don't think those images should have their copyright invalidated.

Re: Database of artists used to train Midjourney AI garners criticism

#65

Post by jncfhnb »

skydhash wrote: post #14
srackey wrote: post #4

I get why artists are trying to stop them, but this battle has already been lost. These tools have been out too long, open source models proliferate freely, and jurisdictions that don’t care about IP law will continue publishing these models. By the time it works it’s way through the courts the situation will be even worse.

Lockpicking tools are widely available, and designs of locks can be found on the web if someone search hard enough. But entering another's home is illegal. It's why businesses pay for licenses even if cracked softwares exists. I hope that artists win these and make using or creating an illegal model an high-risk activity, not worth it for any commercial activity.

Making in image is not like breaking into someone’s house

Re: Database of artists used to train Midjourney AI garners criticism

#66

Post by numpad0 »

bsder wrote: post #17
srackey wrote: post #6

Earlier quoted context omitted.

Completely unenforceable. How can you even tell if an image was made by AI? What if AI created an outline that was worked on by a human artist (or vice versa)? Who would the burden of proof be on? Steam has a “no AI art” policy, and it’s rapidly turning into a “no obvious AI art policy”. How could they tell?

> Completely unenforceable. Complete wrong. You just flip the defaults--something is AI unless you can prove otherwise. This is done already and has precedent. Producing porn requires that you keep artifacts demonstrating that who the performers were, that they were of age, etc. If you claim a work is not AI generated, you should have to produce some artifacts to back up that claim. In the case of a corporation, that…

> something is AI unless you can prove otherwise.

I think this is rather what pro-AI/spammers are trying to do by flooding platforms, that aren't so successful. People don't give as high scores they do for human generated data, and AI images are still considered a form of spam.

Re: Database of artists used to train Midjourney AI garners criticism

#67

Post by rtpg »

srackey wrote: post #6
thurn wrote: post #3

One positive aspect of the status quo in the United States is that AI-generated images are not currently eligible for copyright. I think this is a great direction to go in, I highly doubt Wizards of the Coast or whoever is going to want their premium products to lose copyright protections, so they'll need to keep paying artists. I'd love for us to lean into this -- you can make all the AI art you want, but it automat…

Completely unenforceable. How can you even tell if an image was made by AI? What if AI created an outline that was worked on by a human artist (or vice versa)? Who would the burden of proof be on? Steam has a “no AI art” policy, and it’s rapidly turning into a “no obvious AI art policy”. How could they tell?

"I claim this art was made by this person" "Who?" "OK . did you work on this?" "Where are related work products? Are there any? What about invoices? Simultaneous employment?"

The reality is that most legal things are determined by _convincing people of a truth_. Perhaps you can set up a whole scheme to "launder" AI art and attach names to them. And all the papertrail you generate doing this will show up in discovery in some lawsuit and the copyrights all disappear.

Laws are vibes, not code.

Re: Database of artists used to train Midjourney AI garners criticism

#68

Post by jncfhnb »

tavavex wrote: post #46
kmeisthax wrote: post #19

Earlier quoted context omitted.

I used to agree with this, but there's been research coming out of Google that has altered my opinion. Specifically, Google's gotten rather good at making AI spit out unaltered training set data[0]. This is only possible if the AI is remembering large portions of the original trained-on works, which would make the weights infringing. [0] In the most egregious case, they found that just asking ChatGPT to repeat a word…

I'm kind of confused - you've claimed that Google's AI was broken, but cite an anecdote over ChatGPT? Regardless, even if these cases did happen often enough, it's erroneous to assume that this is something universal (i.e. can be applied to all generative AI models) or intentional. Said models are vastly smaller than the sizes of their training datasets, so it's more or less impossible for all the data to be stored v…

Googles research on chat gpt

Re: Database of artists used to train Midjourney AI garners criticism

#69

Post by doctorpangloss »

vunderba wrote: post #57
srackey wrote: post #4

I get why artists are trying to stop them, but this battle has already been lost. These tools have been out too long, open source models proliferate freely, and jurisdictions that don’t care about IP law will continue publishing these models. By the time it works it’s way through the courts the situation will be even worse.

Yup, and even if the law decides against using copyrighted images as training data, companies such as Adobe are already ahead of the curve with generative systems like firefly which has been trained exclusively on licensed artwork.

It is a misconception that Adobe's models have not been trained on copyrighted work. Nobody should be repeating their marketing claims.

Adobe has not shown how they train the text encoders in Firefly, or what images were used for the text-based conditioning (i.e. "text to image") part of their image generation model. They are almost certainly using CLIP or T5, which are trained on LAION2b, an image dataset with the very problems they are trying to address, C4 (a text dataset similarly encumbered) and similar.

I welcome anyone who works at Adobe to simply answer this question of how they trained the text encoders for text conditioning and put it to rest. There is absolutely nothing sensitive about the issue, unless it exposes them in a lie.

So no chance. I think it's a big fat lie. They'd have to have made some other scientific breakthrough, which they didn't.

Using information from https://openai.com/research/clip and https://github.com/mlfoundations/open_clip, it's possible to investigate the likelihood that using just their stock image dataset, can they make a working text encoder?

It's certainly not impossible, but it's impracticable. On 248m images (roughly the size of Adobe Stock), CLIP gets 37% on ImageNet, and on the 2000m from LAION, it performs 71-80%. And even with 2000m images, CLIP is substantially worse performing than the approach that Imagen uses for "text comprehension," which relies on essentially many billions more images and text tokens.

Re: Database of artists used to train Midjourney AI garners criticism

#70

Post by phantomathkg »

Gigachad wrote: post #63
skydhash wrote: post #14

Earlier quoted context omitted.

Lockpicking tools are widely available, and designs of locks can be found on the web if someone search hard enough. But entering another's home is illegal. It's why businesses pay for licenses even if cracked softwares exists. I hope that artists win these and make using or creating an illegal model an high-risk activity, not worth it for any commercial activity.

The difference is that AI generation tools provide a lot of potential economic value, while breaking in to someones house doesn't. There will be a lot of pressure for governments to allow this to go on since it puts them at advantage over other countries.

> The difference is that AI generation tools provide a lot of potential economic value,

The end does not justify the means.

Post reply on HN