Fat chance. Line must go up.

So, let’s say we create an llm that will be fed will all the copyrighted data and we design it, so that it recalls the originals when asked?! Does that count as piracy or as the kind of legal shananigans openai is doing?