A recent court decision in Thomson Reuters v. Ross Intelligence highlights new legal risks for AI training. Judge Stephanos Bibas ruled that Ross Intelligence directly infringed on Thomson Reuters’ copyrights by using Westlaw headnotes to train its AI-based legal research tool—rejecting Ross’s fair use defense.

This noteworthy ruling addresses non-generative AI and raises questions about how courts may treat generative AI models using copyrighted data. The decision underscores the importance of licensing agreements when training AI tools with proprietary content.

Read our full client alert for key takeaways and legal insights.

For guidance on AI compliance and best practices, contact Robert Feinland at (212) 655-3575 or rf@msf-law.com.