Artist Jingna Zhang built Cara, a platform explicitly intended to protect artists from having their art stolen by AI…and yet these people just won’t leave them well enough alone.
-
Antagonises people who want to stay away from AI
-
Cries when people hate AI
Such a cognitive dissonance
There has to be a word for people, who when they hear someone hates X, but they personally like the X, so they decide to do the X even harder out of spite or as a defiance.
Toddlers, Republicans, Mentally Challenged, Idiots… Lots of words take your pick
-
These are the most deeply insecure people
Anyone know the ways one might remove Glaze and Nightshade? It’d be good to figure out how to make them permanent or show that it’s been tampered with, maybe?
Recently I’ve been embedding metadata into pictures I upload which include the cost to license out the image for AI training: 1 million dollars. Then, if an AI company scrapes my picture, I can take that company to court and have a concrete dollar amount for how much that company owes me for damages.
I think you should also include the text somewhere reachable so no-one can claim that license wasn’t obscured. Its totally fair to strike back with your own type of EULA!
Its a bit delusional to think you can sue them. For a number of reasons, not least of which is that you don’t have the funds for a protracted legal battle and that the burden of proof that your work was used is pretty much impossible to prove.
I don’t think so. First of all, if a lawyer thinks that I have a strong case, they could agree to be paid after I win, so I don’t need to pay out of pocket up front. Second, it is easy to prove that an AI company used my pictures, since they would need to download the pictures to train on them. Alternatively, if a platform sells it to the AI company, there is a record that it was sold for the purposes of AI training.(In that case, we don’t have to prove that the image was used for training, since the crime is the sale.) Lastly, since the license and purchase info is contained within the metadata of the picture, they can’t argue that the license was unavailable or something. They have it right there in their servers!
I’m not a lawyer, but as an artist, bravo. You’ve changed how I look at this. No sarcasm here. 👏🏾👏🏾👏🏾👏🏾
How will you poof that there was this high of a damage?
It’s not for damage, it’s the cost to use the art. If they don’t want to pay for it then they’re free to simply not use it.
I understand that, but you can not simply put random price tags on things and expect that to work in a legal case. Otherwise any pirating of one movie would result in bankruptcy of the person, as the company could just claim 100 fantastillion in damage for one download.
Gross
Yeah good luck asshole.
PSA to anyone who uploads art online, anywhere. Poison it first. Use the latest version. I’m sorry but you are in an arms race now. Dickhead over here is full of shit, but even if they’re not, Nightshade is updated fast.
Poisoning it is sadly an Arms Race, one that will inevitably be lost
And then won, and then lost, and then won. That’s how an arms race works. And we have the advantage in numbers. There are far more people who don’t want their shit scalped than people working for the morons at OpenAI, or Google or Anthropic. And yes, they ARE morons. The battle is only lost when one side gives up. You only lose when you accept defeat.
Unfortunately the anti-AI side is guaranteed to lose eventually, because there’s a definitive lower bound: humans must be able to view and understand the image, ideally with no perceptible loss in quality compared to the original.
This severely limits how far poisoning can go, whereas the theoretical limit for those stealing the art is at least the same level of understanding that humans have (in the sense that our universe physically allows such a mechanism to be implemented, as demonstrated by our bodies).
At the very least it makes the case that the artist believed strongly enough to prevent people from using their art without permission that they use increasingly radical methods.
It makes it so crystal clear that these shitheads are acting in bad faith when they’re circumventing custom made solutions to keep them from using the art.
As if a sign on your website saying “not for use in AI training” shouldn’t be enough. Like imagine I posted art with the message “not for use in movies” and then it got used in a movie. Open and shut.
This. I don’t believe it’s a losing battle like some in this thread, but even if it is, it’s still worth doing.
Yeah, I’m with you there. IMO things like
robots.txtshould have legal weight.
There is no numbers advantage when less than .001% of images are poisoned and of those 99.99% are using outdated tools.
Besides, what’s the goal? Slow its progress down by a few weeks? The inevitable will happen regardless. Do you think people who edited their reddit comments before coming to lemmy made any difference? It’s laughably easy for an AI to tell is garbage data, even easier when you’re using a tool whose source the AI can view.
You know we could just upload an unedited photo to a platform, and then that platform keeps it unedited on a server. Then before exposing to web uses the latest poisoning service periodically. Remove the old edited photo, and apply again to the “raw” with the new updated service. Problem solved. Not super efficient, but still a solution made while shitting on the toilet. I’m sure there’s better.
Yeah fuck this defeatism. Anything you can do you do it. Same thing with anything you strongly disagree with.
Even if you’ve failed don’t give the impression that this was something you went along with. Don’t make it seem like this was just the natural zeitgeist and no one fought it.
Hopefully there will be history in the future, and if there is it should record the fact that there were once humans who tried to be a nuisance over what they believed in even when all the money in the world was against them.
Nightshade and Glaze only run on Windows. Cara has a partnership with them so that your work automatically gets Glazed before being published, hence the anti-AI fame.
Oh cool! Didn’t know that!
“defending AI art”… jesus, that’s sad.
Imagine having so little talent that you consider AI generated slop “art”
Edit*
I hope shitty things like this motivate people to organize into alternative networks, like local meshnets and more in-person community interactions. The internet may be hopelessly broken at this point.
Does anyone have a self hosted social media, where all of the media on my profile would be on my server (even a phone) and somebody viewing my profile would literally have to pull thumbnails and data straight from my device in a peer to peer kind of fashion?
I’ve been keeping an eye on Diaspora, but no clue how good it really is.
It’s amazing how disrespectful these idiots can be. They’ll destroy and steal people’s work, feed it to Big Corpo™ Slopinator 4.7, get shit-tier content in return, just because they chugged the kool-aid that this tech with billions of dubious investments is the future of everything.
Cunt
Actually. I have a hot take.
Forgive me… But the solution is not to replicate Instagram. Which is what Cara does.
Nor DeviantArt, either.
That’s never going to work, for precisely the above reason, and many others.
This is something that’s bothered me immensely about photography (which I’m getting back into as a hobby). Because, for all the “taking a stand” the community does, what do they do en masse?
-
Subscribe to Adobe
-
Live on Instagram
-
Worship corporate influencers.
-
Upload stuff to clouds that the privacy policy says will be used to train.
Just to start.
It infuriates me.
I’m all for the anti-slop stands, but the bulk of the community has precisely zero leg to stand on, if they’re just going to fork over cash hand-and-fist to feed the beasts, and spend all their time on those platforms.
So what do we do?
Make prints. Display them! It can be in museums, festivals, meet ups, in your place, on the street, literally anywhere you want. Even if its on digital screens; as long as its in person, and the digital copy is squirreled away.
And that’s if you feel strongly about the training thing. Personally, I’m more inclined to just post stuff under a permissive license places like Wikipedia could use (albeit not on Instagram).
To be clear, I don’t mean to lump everyone into that. There’s tons of photographers outside that bubble.
…But there are a ton in it, and they happen to be the loudest complainers.
The mass amount of morons wearing pervert glasses will probably just end up copying those prints
Real people are actually wearing those out in the world?
Morons? Yeah
Gross
Not putting the work online slashes accessibility and discoverability though
Yep.
Pick one:
-
Permissive, open visibility.
-
More intimate viewing, restricted usability.
Can’t have one’s cake and eat it.
People thought they could, posting on Instagram or Reddit or wherever, but that was a lie from day one.
You either post online, where people can and will do stuff you don’t like with your content, or you keep it private. This applies to code, writing, communication, everything, not just visual art.
Can’t have one’s cake and eat it.
Sure we can. You don’t need to either use Instagram or never use the net. That’s precisely what Cara is for. Promote open source tooling and use friendly communities that don’t promise to scrape your life’s work.
Now how we deal with mass scrapers downloading images; that’s a legal issue. One that even physical venues isn’t immune to (it just adds a lot of friction).
You either post online, where people can and will do stuff you don’t like with your content, or you keep it private.
I think there’s a difference between “this person took a few images and edited them” and “I scraped an entire website to use for training AI slop”.
you really can’t. If it is online it will be scrapped. You can’t prove it was scrapped, and you can’t stop it. If it is offline it WILL still be scrapped it it just significantly harder to do so.
You can’t prove it was scrapped, and you can’t stop it.
That’s literally why Cara exists. Makes it easier to prove it was scraped and harder for AI to profit off the data it scapes.
Now how we deal with mass scrapers downloading images; that’s a legal issue. One that even physical venues isn’t immune to (it just adds a lot of friction).
That’s the point!
The friction is the key.
If you post images on the internet, how accessible they are directly correlates to how “downloadable” they are. Trying to imitate social media while fighting all scraping is a infinite, costly battle; that’s a service trying to make posts viral and inaccessible. Such a fight takes Facebook money to sustain.
And it’s not necessarily some script kiddie troll that’s boasting about it on Reddit who will eventually get access.
…But add fundamental friction? That raises the bar.
I think the fallacy is artists believe they need an gigantic amalgamation of online people exposed to their art, a large fraciton of the internet. That’s not true. And its a deception, anyway. Smaller niches or smaller communities are fine, and they won’t be such high profile targets.
The friction is the key.
Yeah, and for most people they won’t be bolting their windows and investing in bulletproof glass. A lock is enough detterent. I see sites like Cara as a lock in your metaphor and you’re worried it’s not good enough.
I was thinking in a different dimension and thinking more about community and control. That isn’t something feasible to replicate in the physical world, and its not something we inevitably have to reliquish to billionaire corporations. Community is something you build and cultivate, not something that needs a million dollar investment.
I think the fallacy is artists believe they need an gigantic amalgamation of online people exposed to their art, a large fraciton of the internet.
If you’re a professional, you do. more exposure is more leverage for deals that will pay you. more views correlates with more funding for your paid works, or crowdfunding, or commissions, etc. That’s just how stiff competition is. There’s an old school rule of getting 1000 “true fans” to pay you 100 dollars a year in order to make a good living. Well, we’re in an age where even 10 million impressions may not convert into 1000 true fans. That’s why reach is so important.
Its always been rough being an artist. The kind of rough simply evolved overtime.
-
-
Cara uses third-party content moderation services from Hive (an AI company) to prevent AI-generated art from being uploaded onto the site.
Hahahahaha so this friendly™ Chinese platform basically feeds every uploaded image into an AI to determine if it’s AI.
Just self host your own personal blog/art or use a proper website and not that shitshow.
The idea I believe it to prevent your art from being scraped. Self-hosting Isn’t a solution
The problem is that you presumably want end users to see it, which means a remote machine must be able to download it, which means it inherently must be scrapable (since that’s just downloading and preserving it). You can make it take longer or be more annoying to accomplish, but you can’t actually truly stop it and keep the site usable.
Kind of like how the problem with DRM for video is that the end user can eventually watch it, which means barring all else you can at least point a camera at the screen to produce a copy.
It sucks, yes. But a platform having massive amount of AI-free human created arts, and advertised so, is prime target for scraping. It might actually help the scrapers even.
Unlike many portfolio websites and image sharing platforms, we automatically implement “NoAI” tags on Cara by default to protect images uploaded to the platform. These tags are intended to tell AI scrapers not to scrape from Cara.
Cute of them to think unethical scrapers gonna comply if they ask nicely “dont scwape pwease” The “glaze” or whatever poison they use will get beaten with enough time and training material.
Surely they (cara) have good intentions, but I see this as simply offering an option to artists to instead of getting f* by both scrapers and the platform they are on, the artist can get f* only by scrapers here.
These sites need to find a way to make it easy for bots to scrape a bunch of AI slop images while humans only see human made content.
