826
Nightshade, the free tool that ‘poisons’ AI models, is now available for artists to use
(venturebeat.com)
this post was submitted on 21 Jan 2024
826 points (95.0% liked)
Technology
59374 readers
3392 users here now
This is a most excellent place for technology news and articles.
Our Rules
- Follow the lemmy.world rules.
- Only tech related content.
- Be excellent to each another!
- Mod approved content bots can post up to 10 articles per day.
- Threads asking for personal tech support may be deleted.
- Politics threads may be removed.
- No memes allowed as posts, OK to post as comments.
- Only approved bots from the list below, to ask if your bot can be added please contact us.
- Check for duplicates before posting, duplicates may be removed
Approved Bots
founded 1 year ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
I read that. That's what I've been responding to the whole time. This is a way to analyze and reverse engineer images so you can make your own original works. In the US, the first major case that established reverse engineering as fair use was Sega Enterprises Ltd. v. Accolade, Inc in 1992, and then affirmed in Sony Computer Entertainment, Inc. v. Connectix Corporation in 2000. Do you think SONY or SEGA would have allowed anyone to reverse engineer their stuff if they asked nice? Artists have already said they would deny anyone.
It's not about the data, people having a way to make quality art themselves is an attack on their status, and when asked about generators that didn't use their art, they came out overwhelmingly against with the same condescending and reductive takes they've been using this whole time.
Those are corporations. I'm concerned about how this impacts individuals. Small artists on social media, who make a living off small commissions. I think it is morally and ethically wrong to steal from them.
I also strongly dislike the way you are portraying artists as a monolith. There are some artists who would be willing to submit their art to make an image generation model. You're essentially complaining that not enough people would say yes in your opinion. As though there aren't hundreds of millions of public domain paintings drawings music and all sorts of things that can already be used without screwing over Charlotte and her small time Instagram art dig she affords her 1 bedroom apartment with. You're refusing to even ask her if she's okay with her creations being used in this way.
You're wrong. What you're describing is immoral. I don't care about corporations. They're not who I'm interested in protecting. Its artists themselves. You're also wrong that AI art is some boon for humanity. It's cheap, barely passable noise. Literally, that is what it is. A beefed up toy that mostly exists to generate shitty articles and images that corporations can churn out en mass to manipulate people. That's its best use case at the moment. It's gonna be a very long time, if ever, that human creativity and wit can be engineered.
I don't think images on instagram were used to train open source AI. It's not exactly public. But Charlotte has allowed Meta to train AI on her images.
I think this may be part of the reason why the communication is unsatisfactory. You say you are concerned about the impact on individuals, but what you propose decidedly favors large corporations and the rich. I don't see what difference it makes for Charlotte's life, if an AI is trained on her images. I do see the benefit to Meta and others like it if open AI and smaller start-ups are curtailed.
I don't understand how your proposal would help someone like Charlotte.
Can you explain why you think that requiring people to use public domain or ask for permission to use non-public domain content to train image or text generators would benefit corporations? How does that benefit OpenAI, making them ask before using someones content?
It obviously benefits Meta by hindering the competition and by giving them another source of income.
As to OpenAI, I expect that Microsoft could supply them with quite some user data but I don't think OAI would be a major beneficiary, at first. OpenAI is, after all, a comparatively recent start-up. I think the biggest immediate gains would go to the established Big Tech firms that have their fingers on a lot of user data.
Major content owners, like the NYT, would be able to sell their content again. They have that lying around anyway, so it's pure profit for the owners. Corporations like Getty would also be able to make a killing, not just for being major content owners, but also because they are in the business of ensuring the "provenance" of media. If it's necessary to ask everyone and keep proof on file then they got you covered, for a price.
In the long run, I would expect companies like OpenAI to come out on top. They have their fingers on content generation, so the inequalities would just compound.