Anthropic agrees to $1.5B payout over pirated books

Sep 2025

Anthropic accepted a $1.5 billion settlement with authors who said the Claude team scraped pirate e-book sites to train its chatbot. The deal pays roughly $3,000 per book across 500,000 works, heads off a December trial, and forces one of the richest AI startups to bankroll the writing community it previously treated as free training data.

Incident Details

Perpetrator:AI Vendor

Severity:Catastrophic

Blast Radius:Record copyright settlement drains cash, sets precedent for other AI labs, and fuels public distrust of Anthropic’s data practices.

Tech Stack

Claude chatbotLLM training pipelineWeb-scraped book corpora

References

Los Angeles Times: Anthropic’s $1.5-billion settlement signals new era for AI and artists ↗NPR: Anthropic to pay authors $1.5B to settle lawsuit over pirated chatbot training material ↗PBS/AP: Anthropic to pay authors $1.5B in landmark settlement over pirated chatbot training material ↗