A federal judge has approved a $1.5 billion settlement reached by Anthropic following allegations that copyrighted books were used without permission to train its Claude language model. The decision addresses claims that the AI was trained on pirated literature, potentially