not in any legally reasonable way, and certainly not by anyone who understands how AI (or, really, LLM models) work or what art is.
If it’s not redistributed copyrighted material, it’s not theft
but that’s exactly what OpenAI did-- they used distributed, copyrighted works, used them as training data, and spit out result, some of which even contained word-for-word repetitions of the author’s source material.
AI, unlike a human, cannot create unique works of art. it can old produce an algorithmically-derived malange of its source-data recomposited in novel forms, but nothing resembling the truly unique creative process of a living human. Sadly, too many people simply lack the ability to comprehend the difference.