'Uneasy Feeling With All of This Money on the Table': Judge Blasts Anthropic's $1.5 Billion Book Copyright Settlement Federal Judge William Alsup denied a deal between AI company Anthropic and book authors over copyright infringement claims.
By David James
Key Takeaways
- A proposed $1.5 billion copyright settlement between AI company Anthropic and book authors represented in a class action lawsuit has been denied in federal court.
- The judge cited several issues that left him feeling "uneasy" about the proposal.
- The settlement will likely provide a template for future AI copyright infringement claims.
A federal judge has cast doubt on the proposed $1.5 billion copyright settlement between AI company Anthropic and book authors represented in a class action lawsuit, delaying its approval.
Judge William Alsup declined to approve the settlement on Monday, citing that authors might be excluded from meaningful input as negotiations unfolded behind closed doors.
"I have an uneasy feeling about hangers-on with all this money on the table," Alsup said, per Bloomberg.
Related: Anthropic Is Now One of the Most Valuable Startups of All Time: 'Exponential Growth'
Alsup noted the settlement was "nowhere close to complete" and required further clarification on vital aspects, including how claims would be filed, how class members would be notified, and which works were covered. Without these, Alsup argued, the deal could unfairly disadvantage authors and lead to future litigation.
The lawsuit originates from Anthropic's alleged downloading of millions of copyrighted books to train its AI models—a claim echoing similar legal efforts against major tech firms like OpenAI and Meta. Anthropic proposed paying about $3,000 per book to 500,000 authors in the suit.
Alsup said there were many "important questions" that need answering before approving the settlement, including a complete list of books and a clearly defined process for notifying potential class members, adding that class members typically "get the shaft" once deals are made and "attorneys stop caring." He wants clear and early guidance provided to authors, giving them proper time to opt in or out of the suit.
All legal eyes are on what happens next as this suit is thought to provide a template for future AI copyright litigation.
Related: 'We Don't Negotiate': Why Anthropic CEO Is Refusing to Match Meta's Massive 9-Figure Pay Offers
