Breaking

US Judge Grants Final Approval to Anthropic's $1.5 Billion Copyright Settlement With Authors

July 21, 2026 at 00:01 EDT

Policy & Regulation
Foundation Models

A US federal judge has granted final approval to a $1.5 billion settlement resolving claims that Anthropic used pirated books to train its Claude AI models, in what is being described as the largest copyright payout on record. The decision closes a landmark case that has become a reference point for how the courts weigh generative AI training against the rights of authors and publishers.

Bartz et al. v. Anthropic PBC · N.D. Cal. 3:24-cv-5417

Judge Approves $1.5 Billion Settlement Over Pirated Books Used to Train Claude

A US federal judge has granted final approval to what is being described as the largest copyright payout on record — resolving claims that Anthropic downloaded millions of pirated books to train its AI models.

$1.5B

Total settlement — among the largest copyright recoveries in US history

~500K

Eligible works covered, after duplicates are removed

~$3,000

Per work paid to rights holders, after costs & fees

7M+

Titles alleged downloaded from pirate repositories

From alleged infringement to author payouts

Scale of the works involved (proportional)

7M+

Titles allegedly downloaded & stored

~500K

Eligible works in the settlement class

The court's split reasoning

Training may be transformative — but how the data was acquired is a distinct legal exposure.

Training on lawful works

Potentially transformative — may qualify as fair use.

Downloading pirated copies

Separate liability — regardless of eventual use.

Validation for authors

Affirms that training data must be lawfully acquired, and gives affected rights holders a concrete per-work recovery. Anthropic must also destroy the pirated copies it obtained.

Still unsettled

The deal resolves only the piracy question and covers past infringement — not future use. The broader fair-use status of AI training remains open, and future cases could reach different conclusions.

Continue reading

The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.