Nvidia Fails to Escape Authors’ AI Class Action Copyright Claims

May 6, 2026, 5:58 PM UTC

Chipmaker Nvidia Corp. must face claims it trained its NeMo Megatron large language model on copyrighted books in a proposed class action brought by a trio of authors.

Abdi Nazemian, Brian Keene, and Stewart O’Nan plausibly alleged Nvidia trained Megatron 345M on a dataset that included their books, the US District Court for the Northern District of California said. The authors also preserved their claims for contributory infringement, but Judge Jon S. Tigar dismissed vicarious infringement claims in his Tuesday ruling without prejudice.

Tigar also denied bids to dismiss claims regarding allegations Nvidia infringed using “unidentified models” other than Megatron ...

Learn more about Bloomberg Law or Log In to keep reading:

See Breaking News in Context

Bloomberg Law provides trusted coverage of current events enhanced with legal analysis.

Already a subscriber?

Log in to keep reading or access research tools and resources.