Federal Judge Delays Anthropic’s $1.5B Copyright Settlement Approval

A federal judge has halted the approval of Anthropic's historic $1.5 billion copyright settlement, citing unresolved issues regarding attorney fees and payments to lead plaintiffs. This decision leaves approximately 120,000 authors and copyright holders waiting for clarification on a case that has been in motion since 2024.

U.S. District Judge Araceli Martinez-Olguin raised several questions during a hearing on May 14 in San Francisco that remain unanswered. The case arises from allegations that Anthropic illegally downloaded over 7 million books from shadow libraries, including LibGen and PiLiMi, to train its artificial intelligence models.

The lawsuit's origins trace back to a June 2025 ruling by now-retired Judge William Alsup, who deemed Anthropic’s use of these materials as fair use, provided they were legally acquired. However, he pointed out that gathering pirated copies in a “central library” for purposes beyond training did not fall under fair use. This led to a potential trial set for December 2025, where damages could reach hundreds of billions of dollars. To avoid this trial, Anthropic agreed to a settlement that Alsup initially approved in September 2025.

After Alsup's retirement, the case was reassigned to Martinez-Olguin, who is now responsible for finalizing the settlement terms. According to court filings, authors and publishers have submitted claims covering more than 91% of the roughly 480,000 eligible works in the settlement class. This participation rate sharply contrasts with the 9% median seen in typical U.S. consumer class actions, as reported by the Federal Trade Commission in 2019.

Justin Nelson, the lead plaintiff attorney representing the authors, emphasized the significance of the claims rate, stating, “This claims rate is another reason why this settlement is so historic and demonstrates the overwhelming support of the class.” Under the proposed settlement, each title is expected to receive a minimum of $3,000, which will be shared between authors and publishers. Notably, self-published authors and those whose rights have reverted will retain the full amount.

The current uncertainty surrounding attorney fees complicates an already intricate case, as stakeholders await more definitive guidance from the court. The outcome of this settlement could have significant implications for copyright law and the use of proprietary works in training AI systems, especially as the industry navigates the tension between innovation and intellectual property rights.

As the case develops, the implications for both Anthropic and the broader AI sector remain uncertain. With the judge's request for clarity on fees, the timeline for resolution may extend further, leaving authors and copyright holders in anticipation of their future compensation and the legal precedents that may emerge from this landmark case.