Anthropic’s $1.5B copyright settlement approved; only 350 authors opted out (arstechnica.com)

A U.S. District Judge has approved a landmark $1.5 billion settlement between Anthropic and a group of authors, marking the largest copyright class-action settlement in history. This resolution follows a contentious debate over Anthropic’s use of copyrighted books to train its AI, where the court highlighted that while training could align with fair use, the potential piracy of individual works was not justified. Despite some authors protesting the settlement's terms, claiming inadequate payouts and excessive legal fees, nearly 91% of eligible authors supported the settlement, with only a small number opting out or objecting. The decision is significant for the AI/ML community as it sets a precedent for how copyright laws may adapt to the evolving landscape of AI training. Judge Araceli Martínez-Olguín emphasized that the settlement provided each author with approximately $3,000 per work – significantly higher than minimum statutory damages – reinforcing the notion of fair compensation. Furthermore, the judge reduced attorney fees from 20% to about 7% of the total settlement, ensuring that more funds will flow directly to the impacted authors, which may influence future negotiations regarding AI training and intellectual property rights in the rapidly advancing field of artificial intelligence.