Anthropic to Pay CZK 33.6B to Authors for AI Training

Anthropic, the developer of the popular Claude AI chatbot, reached a record settlement of $1.5 billion (about CZK 33.6B) with a class of authors who sued over the unauthorized use of their books for AI training. This historic settlement is the first of its kind in the generative AI era and sets a precedent for the entire industry.

Key settlement terms

Category	Detail
Total amount	$1.5 billion (~CZK 33.6B)
Per‑book payout	$3,000 for each of an estimated 500,000 books
Comparison to typical standards	4× higher than statutory minimums, 15× higher than compensation for “innocent” infringement
Anthropic obligations	- Delete pirated datasets from Library Genesis and Pirate Library Mirror - Four staged payments, first $300M within 5 business days - If the number of works exceeds 500,000, an additional $3,000 per extra work

Background of the dispute

Original lawsuit

A trio of authors — Andrea Bartz (thriller author), Charles Graeber, and Kirk Wallace Johnson (journalists and nonfiction authors) — filed suit in 2024 on behalf of a broader group of writers and publishers. They accused Anthropic of:

• Massive copyright infringement by downloading millions of books from pirated “shadow libraries”
• Commercial use without permission to train the Claude model
• Systematic piracy via Library Genesis and Pirate Library Mirror

Why did Anthropic settle?

Existential financial risk: U.S. copyright law allows statutory damages up to $150,000 per willful infringement. With more than 7 million pirated books, Anthropic faced potential damages in the hundreds of billions of dollars. Class‑action pressure: Anthropic warned the courts that certifying a class action creates “undue pressure” to settle and a “death‑knell” situation that could force an unfair settlement regardless of the case’s merits. Strategic reassessment: The case was scheduled for trial in December 2025, with potential damages in the hundreds of billions of dollars.

Impact on the AI industry

Licensing precedent: “This settlement marks the beginning of a necessary evolution toward a legitimate, market‑based licensing system for training data,” said attorney Cecilia Ziniti. Message to competitors: “This settlement sends a powerful signal to AI companies and creators alike that stealing copyrighted works from pirate sites is wrong,” said plaintiffs’ attorney Justin Nelson. Financial strength of AI firms: Anthropic is in a strong position to absorb significant compensation. On Tuesday the company announced a new $13B funding round, bringing its valuation to $183B.