Federal Judge Rules AI Training as Fair Use, Anthropic Faces Piracy Claims
Landmark Ruling on AI Training and Copyright
In a groundbreaking decision, U.S. District Judge William Alsup ruled that Anthropic's use of published books to train its artificial intelligence models without explicit author permission constitutes fair use under copyright law. This marks the first judicial endorsement of an AI company's fair use defense in training data disputes.
Industry-Wide Implications
The ruling carries significant consequences for dozens of pending lawsuits against major tech companies including OpenAI, Meta, and Google. While not binding on other courts, Judge Alsup's interpretation provides a legal framework that may favor technology companies in similar copyright battles.
At the heart of these cases lies interpretation of the fair use doctrine - a provision of U.S. copyright law last updated in 1976, long before the advent of generative AI. Courts typically evaluate four factors in fair use determinations:
- The purpose and character of the use
- The nature of the copyrighted work
- The amount used in relation to the whole work
- The effect on the potential market
Judge Alsup found that Anthropic's use of literary works for AI training constituted transformative use, a key consideration under fair use doctrine.
Piracy Allegations Remain Unresolved
While Anthropic prevailed on the training data issue, the court left open serious questions about how the company obtained some materials. In Bats v. Anthropic, plaintiffs allege the company:
- Attempted to create a "central library" containing "all books in the world"
- Downloaded millions of copyrighted books from piracy websites
- Later purchased some previously pirated works
The judge noted: "We will address the pirated books used by Anthropic to create the central library and the damages they caused...Anthropic later purchased books that had previously been stolen online, which does not absolve it of theft liability."
The case now moves to discovery phase regarding these piracy claims, with potential statutory damages still in play.
Key Points:
- First judicial ruling supporting AI training as fair use under copyright law
- Decision may influence dozens of similar cases against tech companies
- Fair use doctrine interpretation becomes central to AI copyright battles
- Anthropic still faces serious piracy allegations regarding source materials
- Case continues with discovery phase on potential damages