As Meta defends its AI training practices in court, major publishers and copyright law experts are weighing in against the company. An amicus brief from publishers highlights Meta’s alleged reliance on pirated book archives including Anna’s Archive and Z-Library. Separately, a brief from law professors argues that Meta’s unauthorized copying to train Llama is an “undeniably commercial” use that provides no new transformative meaning and shouldn’t qualify as fair use.
No, not least because almost nothing in this area is self-evident due to the state of caselaw at the moment.
Putting aside for the moment the question of whether “generative” implies “transformative” in the specific sense under discussion in copyright law, the definition of “transformative” in this context is highly contentious, and courts have avoided defining it in an unambiguous way. Even here, the courts will probably avoid answering these questions if at all practical.
This is a big part of why fair use is in such a bad state right now: no predictability in how courts will rule on it as a defense, plus no way to keep you out of courts in the first place.
That’s not completely true. Fair Use has four declarations that you must pass, so there’s at least some definition to how it would play out in the courts. But, it’s not a definitive rule of law, so yeah it’s not going to keep you out of the courts.