A $1.5B copyright win for authors that might backfire
anthropic approved a record $1.5m copyright settlement over claude training data, setting a precedent that could stifle ai development by making datasets legally radioactive.
anthropic agreed to pay $1.5 billion to settle a class-action lawsuit alleging it used pirated books to train its claude models, marking the largest known copyright payout in ai history. plaintiffs successfully argued that the company profited from copyrighted works without permission, forcing a settlement that includes substantial compensation. this legal victory emboldens creators to demand compensation but raises alarms that treating training data as infringing could make open-source model development legally impossible. industry watchers expect this deal to trigger a wave of new claims against other major labs.