A federal judge granted final approval of a $1.5 billion class-action settlement between Anthropic and a class of authors and publishers whose copyrighted books were allegedly sourced from pirated libraries LibGen and PiLiMi to train Anthropic’s Claude models. The settlement covers roughly 500,000 works, pays affected rightsholders about $3,000 per work, and requires Anthropic to destroy any pirated datasets still in its possession. It is the largest copyright class-action settlement in history and the first major court resolution of an AI training-data copyright dispute.