Appeals court rejects fair use defence in first AI‑training copyright ruling
The 3rd US Circuit has upheld Thomson Reuters' win over Ross Intelligence, the first appeals-court ruling on whether training an AI system on copyrighted material is fair use. Its reasoning is sealed.
The Philadelphia-based 3rd US Circuit Court of Appeals on Tuesday upheld a lower-court ruling for Thomson Reuters in its copyright dispute with the former legal-research rival Ross Intelligence, Reuters reported. The court rejected Ross' argument that its search engine made fair use of material from Thomson Reuters' Westlaw platform. It is the first ruling of its kind from a US appeals court.
Thomson Reuters sued in 2020, accusing Ross of copying Westlaw "headnotes", short summaries of points of law in court opinions, and using thousands of them to train a competing AI-based legal search engine. According to MediaPost, Thomson Reuters alleged Ross "surreptitiously" obtained the material from a licensee to build a rival product without bearing the development cost.
Unlike most of the AI copyright fights now in court, the case does not involve generative AI, the kind that produces new text or images, Reuters noted. It is nonetheless the first AI-training dispute to reach an appeals court. The district court had rejected Ross' fair use defence in February 2025, finding it used the headnotes for the same commercial purpose as Westlaw.
The appeals court's reasoning is not yet public. MediaPost reported the opinion is temporarily sealed pending decisions on confidentiality, so the extent to which the panel followed the lower court cannot yet be checked. Thomson Reuters said it believes respecting copyright is essential to innovation. Ross, which shut its platform in 2021 citing litigation costs, did not respond to Reuters' request for comment.