Anthropic asks for summary judgment in Concord Music Group Inc. v. Anthropic PBC, arguing that training Claude on song lyrics is transformative and protected by fair use, and that restricting transformative AI would run counter to the Copyright Act’s public-serving purpose.

The company contends that market dilution for AI-generated works is generic and not a direct substitution of specific lyrics, thereby weakening the fair-use barrier on market-dilution grounds.

Anthropic emphasizes that the fair-use defense should not hinge on limiting transformative AI technology, highlighting the broader public-policy goals of copyright law.

Publishers and Anthropic offered no immediate comments on the filing.

The dispute began with lawsuits against Anthropic in 2023 from UMG and other publishers, with additional actions by major label groups in 2024; while some licensing settlements have occurred, several cases remain active.

This case fits within a growing pattern of AI-training copyright settlements and decisions, including a prior $1.5 billion authors’ settlement involving a competing group last year.

Earlier actions by publishers and creators show divergent rulings as courts assess fair use in AI training, with notable settlements and ongoing litigation around authors and OpenAI-related entities.

Publishers have not publicly commented on the filing; courts will decide whether copying millions of copyrighted works for AI training constitutes fair use.

Coverage on fair use in AI debates frames this case within a broader landscape of related litigation across tech and content companies.

The fair-use question remains unsettled and central to multiple AI copyright cases against major tech players, including actions against other AI training platforms.

This case is Concord Music Group Inc v. Anthropic PBC, filed in the Northern District of California, No. 5:24-cv-03811.

The matter is before Judge Eumi Lee in the U.S. District Court for the Northern District of California.