I don't really understand why we're more strict over here about it and I'd just like some feedback on why that is the case. I had my first two posts here deleted under that rule and I just didn't expect to run into it at all.

throwawayvr said:
I don't really understand why we're more strict over here about it and I'd just like some feedback on why that is the case. I had my first two posts here deleted under that rule and I just didn't expect to run into it at all.

I haven't been told the reason, but my guess is that it is to avoid legal complications that could be caused by image models that are trained on actual child pornography. Even if 100% of those posts come from models trained only on furry drawings, the risk remains.

The legal battles involving AI art are just starting. For all we know this site will have to shut down within a few years if the copyright battles play out a certain way.

lance_armstrong said:
I haven't been told the reason, but my guess is that it is to avoid legal complications that could be caused by image models that are trained on actual child pornography. Even if 100% of those posts come from models trained only on furry drawings, the risk remains.

The legal battles involving AI art are just starting. For all we know this site will have to shut down within a few years if the copyright battles play out a certain way.

I don't really understand how the risk remains if 100% came from furry drawing training? Also isn't that risk there even when generating art that *isn't* underage?

In any case, thank you for answering me even if the answer is a guess, I do appreciate it.

Here's to a long life of the site (and a hope that the rules can change someday!)

throwawayvr said:
I don't really understand how the risk remains if 100% came from furry drawing training? Also isn't that risk there even when generating art that *isn't* underage?

In any case, thank you for answering me even if the answer is a guess, I do appreciate it.

Here's to a long life of the site (and a hope that the rules can change someday!)

Public AI image models aren't trained on illegal content (or at least, if they are, they're good at hiding it), but the problem is, people can train their own incarnations of Stable Diffusion locally.