4 ms·

While this may very well be covered, I think the general problem in meatspace is that there was no advance notice given to exercise the option to place the prop

While this may very well be covered, I think the general problem in meatspace is that there was no advance notice given to exercise the option to place the proper robots.txt directives to opt out of having one’s artwork collected for model training before it happened, while still preserving the ability to have one’s artwork findable by search engines and the like. I’m sure there are more than a handful of people who have no idea that a robots.txt file can be used to prevent AI data collection—and some may even be surprised to learn the file that’s been used for search engine crawlers is also going to double for AI crawlers. To push a bit further, there’s something that just feels particularly off about assuming everyone’s content is up for grabs unless the producers do the work to opt out. I think there’s an especially palpable bit of irony looking at it from the EU’s perspective—where cookies must be opt-in, but grabbing all your copyrighted material so companies can do whatever they like with it places the burden on the owner to opt-out. It just feels backward. Perhaps one should have to expressly opt-in to allowing their work to be accessible as training data. At least then there will be a clear signal that the producer of the work can’t later complain, as they willingly granted permission.

Karunamon 3y ago
I wonder if these authors would have complained so loudly if they had known that other artists were looking at their output to learn how to create their own work? Absolutely none of them sprung from the womb, tablet in hand, to create their work ex nihilo, based on nothing other than their own entirely original thoughts.
astrange 3y ago
Their art wasn't collected for model training. The #1 artist supposedly being copied in SD1.5 is not in the training data. Artists just don't know how the model works and think you need to put in an "art" image to get an "art" image out, but of course that's not true.
- grumbel 3y ago
  > and think you need to put in an "art" image to get an "art" image out, but of course that's not While you don't need that to generate an image, it's something SD can actual do extremely well with ControlNet, Textual Inversion, LoRA, img2img and so on. That's an area where things are going to get interesting in the future, as you can take any image, feed it into SD and produce hundreds of AI images from it. Very easily, without much effort and within minutes. The delineating line between original work and derivative becomes extremely blurry here, as what you are copying is not "the image", but just concepts within the image, that can be a pose, camera angle, scene layout, art style or really anything. You can "copy" it with as much variation as you want, you can remix it with other images, text prompts and so on. Where does "looking at reference" stop and "doing a copyright violation" start? The spooky part with AI art that it stops images from being singular entities, with AI you can explode every piece into millions of possible variations. AI is so fast at generating art that a future where we could generate movies in real time might not be far away. It's already fast enough to produce images and text stories faster than you can consume them. There might be a fundamental shift in art consumption ahead of us.

Karunamon 3y ago

I wonder if these authors would have complained so loudly if they had known that other artists were looking at their output to learn how to create their own work? Absolutely none of them sprung from the womb, tablet in hand, to create their work ex nihilo, based on nothing other than their own entirely original thoughts.

astrange 3y ago

Their art wasn't collected for model training. The #1 artist supposedly being copied in SD1.5 is not in the training data. Artists just don't know how the model works and think you need to put in an "art" image to get an "art" image out, but of course that's not true.

grumbel 3y ago
> and think you need to put in an "art" image to get an "art" image out, but of course that's not While you don't need that to generate an image, it's something SD can actual do extremely well with ControlNet, Textual Inversion, LoRA, img2img and so on. That's an area where things are going to get interesting in the future, as you can take any image, feed it into SD and produce hundreds of AI images from it. Very easily, without much effort and within minutes. The delineating line between original work and derivative becomes extremely blurry here, as what you are copying is not "the image", but just concepts within the image, that can be a pose, camera angle, scene layout, art style or really anything. You can "copy" it with as much variation as you want, you can remix it with other images, text prompts and so on. Where does "looking at reference" stop and "doing a copyright violation" start? The spooky part with AI art that it stops images from being singular entities, with AI you can explode every piece into millions of possible variations. AI is so fast at generating art that a future where we could generate movies in real time might not be far away. It's already fast enough to produce images and text stories faster than you can consume them. There might be a fundamental shift in art consumption ahead of us.