Judge Approves $1.5 Billion Anthropic Deal
A US federal judge has approved a $1.5 billion settlement between artificial intelligence company Anthropic and authors who accused it of unlawfully acquiring pirated copies of their books. The agreement is considered the largest known copyright recovery in US history.
Anthropic Copyright Settlement Approved
US District Judge Araceli Martínez-Olguín granted final approval to the class-action settlement in San Francisco. The lawsuit was filed in 2024 by authors Andrea Bartz, Charles Graeber and Kirk Wallace Johnson.
The writers alleged that Anthropic downloaded millions of pirated books while building a central digital library used during the development of its Claude artificial intelligence models. Anthropic is backed by major technology companies, including Amazon and Alphabet.
Authors to Receive Compensation
The settlement covers more than 482,000 eligible books, with over 91% already claimed by authors or publishers. Eligible copyright holders are expected to receive approximately $3,000 for each affected work, although individual payments may vary.
Anthropic must also destroy the original files obtained from pirate websites and any copies created from them. The court approved more than $101 million in legal fees for the lawyers representing the authors.
Some writers rejected the settlement after arguing that the proposed payment was insufficient. They have opted to pursue separate copyright lawsuits against Anthropic and other AI companies.
AI Training and Copyright Law
An earlier ruling made an important distinction between AI training and the unlawful acquisition of copyrighted material. The court found that training AI systems using lawfully obtained books could qualify as fair use.
However, maintaining a permanent digital library containing millions of pirated books was found to be copyright infringement. The settlement resolves claims relating to Anthropic’s past acquisition and storage of those works but does not grant permission for their future use.
The decision could influence several pending copyright cases involving AI developers, authors and publishers.
