Meta’s AI copyright trial set for May 2027 as works list remains unsettled

A US federal court has scheduled a May 3, 2027, trial in Entrepreneur Media’s copyright lawsuit against Meta, as the parties continue to identify works allegedly used in AI training.

Meta Platforms is set to face a copyright trial on May 3, 2027, in litigation brought by Entrepreneur Media over the alleged use of copyrighted books and other works to train its artificial intelligence (AI) models.

The trial date was confirmed in a joint case management statement filed October 27 in the US District Court for the Northern District of California, following an October 14 order by Judge Vince Chhabria advancing the proceedings from May 24, 2027.

The case is one of several copyright actions being managed alongside litigation brought by authors and publishers against Meta over its use of books to train its Llama AI models.

The court has indicated that the May 3 trial slot may be used by another related case if Entrepreneur Media settles.

Copyright works remain under scrutiny

The litigation concerns allegations that Meta used copyrighted books obtained from unauthorized sources to develop its AI systems. The precise list of works at issue remains unsettled. Entrepreneur Media is expected to seek permission to amend its complaint to add more than 400 books and over 140 magazine issues identified through Meta’s discovery responses and related evidence.

The parties continue to dispute aspects of how certain materials were obtained and used. The case management statement addresses allegations concerning the Books3 dataset, a collection of books plaintiffs say was obtained through torrenting. Meta disputes aspects of the plaintiffs’ account, including the circumstances surrounding the dataset’s acquisition and whether particular downloads were connected to the company.

The dispute is part of a broader wave of US litigation examining the use of copyrighted material in generative AI development, including questions surrounding unauthorized acquisition, distribution, and the applicability of copyright defenses.

Key milestone in AI copyright litigation

The May 2027 trial date provides a concrete milestone in a closely watched dispute over the relationship between copyright law and AI development. While other AI copyright cases have focused heavily on whether training constitutes fair use, the current proceedings also place emphasis on how training materials were acquired and whether copyrighted works were distributed without authorization.

The court has scheduled hearings on dispositive motions for February 25, 2027, ahead of the trial. A pretrial conference is currently set for April 20, 2027, although the parties have identified a potential scheduling conflict that could prompt a change. The outcome could have implications for rights holders seeking to establish how copyrighted works were used in AI training, as well as for developers facing questions about the provenance of their training datasets.

With the list of works still being finalized, the case is expected to remain an important development in the evolving US legal landscape surrounding generative AI and copyright.

← Back to News