Anthropic’s $1.5B settlement with authors approved
Anthropic agreed to pay $1.5 billion to authors whose pirated books it used to train its AI, marking the largest U.S. copyright settlement.Summary
A federal judge in Oakland approved Anthropic’s $1.5 billion settlement with more than 500,000 authors and publishers who sued the company for using pirated books to train its AI models. The settlement, the largest copyright case in U.S. history, compensates authors roughly $3,000 per eligible work and awards $101.6 million in legal fees to the plaintiffs’ lawyers. While the court found that training an AI on copyrighted material can be fair use, it ruled that Anthropic’s acquisition of the books—by downloading millions from pirated libraries and converting them to digital copies—was unlawful. The decision sets a precedent for future AI‑related copyright disputes and clarifies that the legality of AI training hinges on how the source material is obtained, not merely on the training process itself.
(Source：Columbian)