Intelligence Piracy: Anthropic Agrees to US$1.5 Billion Copyright Settlement in Landmark AI Case

Artificial intelligence developer Anthropic has received final court approval for a landmark US$1.5 billion settlement resolving copyright claims brought by a class of authors whose books were allegedly copied without permission to build a digital library used in connection with training its Claude AI models.

The settlement, approved by a U.S. federal judge, is the largest copyright agreement of its kind in U.S. history and marks another significant development in the growing number of legal disputes between rights holders and AI developers.

Earlier in the proceedings, the court distinguished between the act of training AI models and the way copyrighted works were acquired. The judge held that training an AI model using lawfully obtained books could qualify as fair use because the process was sufficiently transformative. However, the court found that the alleged acquisition and storage of millions of pirated books created potential copyright liability, leading to the settlement.

Court documents indicate that more than 91% of eligible authors and publishers have submitted claims for compensation, with average payments expected to be approximately US$3,000 per qualifying work. The judge also reduced the legal fees sought by class counsel, concluding that a percentage-based award would have unnecessarily reduced the funds available to rights holders.

For those working in anti-piracy and intellectual property enforcement, the case is another reminder that the source of digital content remains a critical part of any investigation. As AI companies continue to face scrutiny over the datasets used to develop their models, the provenance of copyrighted material is becoming an increasingly important area of intelligence gathering and enforcement.

The Anthropic litigation is one of several ongoing cases expected to shape the legal boundaries of AI training and copyright over the coming years, with further decisions likely to have significant implications for technology companies, publishers and rights holders worldwide.

References

1. Reuters. US judge approves Anthropic’s $1.5 billion settlement of copyright lawsuit. (20 July 2026) https://www.reuters.com/world/us-judge-approves-anthropics-15-billion-settlement-copyright-lawsuit-2026-07-20/  

2. Reuters. In landmark Anthropic settlement, judge rejects ‘windfall’ for lawyers. (21 July 2026) https://www.reuters.com/legal/government/landmark-anthropic-settlement-judge-rejects-windfall-lawyers-2026-07-21/  

3. Associated Press. Judge approves a $1.5B Anthropic settlement over pirated books used to train the Claude chatbot. (21 July 2026)

   https://apnews.com/article/74b140444023898aeba8579b6e9f0d63  

4. The Verge. Anthropic’s $1.5 billion book piracy settlement approved by judge. (21 July 2026) https://www.theverge.com/ai-artificial-intelligence/968724/anthropic-authors-settlement-ai-copyright-approved