Whoa. "[F]inetuning exclusively on Haruki Murakami's novels unlocks verbatim recall of copyrighted books from over 30 unrelated authors...Our findings offer compelling evidence that model weights store copies of copyrighted works."
papers.ssrn.com/sol3/papers....
Frontier LLM companies have repeatedly assured courts and regulators that their models do not store copies of training data. They further rely on safety alignme
