I think they were talking about the "model giants" like OpenAI you mentioned. Not saying they're correct, but I will concede the amount of copyrighted information someone like OpenAI would want is probably (at least) an order of magnitude more than this particular case.
No comments yet.