Anthropic Settles Copyright Lawsuit with Authors for $1.5 Billion
Key Points
- Anthropic agreed to a $1.5 billion settlement with authors.
- Approximately 500,000 authors will receive about $3,000 per work.
- The case centered on books downloaded from Library Genesis (LibGen).
- Anthropic will delete all illegally obtained for training.
- The settlement ends remaining legacy piracy claims.
- Court previously ruled Anthropic’s overall training approach qualifies as fair use.
- Lawyer for authors called the deal the first of its kind in the AI era.
- Anthropic maintains it did not use pirated works for publicly released AI.
Anthropic, the creator of the Claude chatbot, agreed to a $1.5 billion settlement to resolve a class‑action copyright lawsuit brought by authors whose works were downloaded from piracy sites such as Library Genesis. The deal will provide roughly $3,000 per work to about 500,000 authors and requires Anthropic to delete the illegally obtained material. While the company maintains it did not use pirated works for publicly released AI, the settlement ends the remaining legacy claims and marks a landmark moment for AI‑related copyright disputes.
Background
Anthropic, known for its Claude chatbot, faced a class‑action lawsuit alleging that the company used copyrighted books that were downloaded from illegal sources, notably Library Genesis (also called "LibGen"). The plaintiffs, representing roughly 500,000 authors, argued that Anthropic’s training data included pirated works, violating copyright law.
Settlement Details
The settlement amounts to $1.5 billion, the largest payout ever recorded for a U.S. copyright case. Under the agreement, each author will receive about $3,000 per work. Anthropic also agreed to delete all content that was obtained illegally and affirmed that it did not use any pirated material to develop AI technologies that have been publicly released. The company has not admitted wrongdoing.
Legal Context
Earlier, a district court ruled that Anthropic’s overall approach to training its large language model qualified as fair use, a decision that the company highlighted as a significant victory. However, the same judge allowed the authors and publishers to pursue piracy claims because the source material had been downloaded from illicit sites. The settlement resolves those remaining legacy claims.
Reactions
Representing the authors, lawyer Justin A. Nelson described the settlement as “the first of its kind in the AI era” and emphasized its historic scale. Anthropic’s Deputy General Counsel Aparna Sridhar reiterated the company’s commitment to safe AI development while noting that the settlement resolves the plaintiffs’ remaining claims.
Implications
The agreement sends a clear signal to AI developers about the legal risks of using copyrighted material obtained from illegal sources. It also establishes a precedent for sizable compensation to creators when their works are used without permission, even as courts continue to grapple with the boundaries of fair use in AI training.