The settlement centers on copyright infringement tied to unlawfully obtained training materials used for AI models.

A class-action settlement in Bartz v. Anthropic PBC approved by the Northern District of California creates a $1.5 billion settlement fund.

The case is described as potentially the largest settlement of its kind for an AI company to date.

The settlement was approved in mid-summer 2026 and was reported in October 2026, highlighting ongoing legal scrutiny of AI training data practices.

The report underscores the risk to AI companies from copyright claims related to training data, even when model training itself may be considered fair use.

The court’s stance indicates that AI training may qualify as fair use in some contexts, but liability remains for improper sourcing of training data.