⚖️ Anthropic to Pay $1.5 Billion for Illegally Downloaded Books

Anthropic has agreed to a $1.5 billion preliminary settlement in what could become the most significant digital copyright case in U.S. history. The deal still requires judicial approval.

Anthropic is accused of torrenting more than 7 million books, in addition to scanning legally purchased editions, to train its Claude model.

The judge ruled that training AI on lawfully acquired books qualifies as fair use, but using pirated copies is a direct copyright violation.

Notably, Anthropic will compensate authors for only 500,000 works (at $3,000 each), which fall under the formal criteria of U.S. law.

💰 Why Settle?

Anthropic is seeking to "cut its losses." Cornell Law professor James Grimmelmann called the $1.5 billion deal a "modest settlement" compared to potential damages of more than $10 billion if a jury sided with the authors.

Despite accepting the settlement, the startup does not admit liability and insists that pirated copies were not used in building its commercial AI models.

🌐 What It Means for the Industry

Similar lawsuits have already been filed against Google, Meta, Microsoft, and OpenAI—and the Anthropic case could set a precedent for future litigation.

#Anthropic #law @hiaimediaen