Wire · regulatory
Judge approves a $1.5B Anthropic settlement over pirated books used to train the Claude chatbot
◆ Sectors
◆ Geography
◆ Source
◆ Verified
Fusion42 · 21 July 2026 · Fusion42 review
A federal judge approved a $1.5 billion copyright settlement against Anthropic for using pirated books to train Claude, with about 91% of 482,000 affected authors and publishers now eligible to claim roughly $3,000 per book.
This Wire brief sits within Fusion42's coverage of AI Frontier Models, and 5 sources have reported it.
◆ ◆ The Wire takeaway
If you're training AI models on unlicensed text, Anthropic just set your cost of doing so at roughly $6 per book across a dataset. The court found training on copyrighted material itself isn't illegal—but acquiring it through piracy is, and the bill is now fixed and visible to every other AI company watching this case.
◆ Coverage
5 sources · 21 Jul 2026
◆ Related on Wire
◆ Topics