A US federal judge has ruled that Ross Intelligence infringed on Thomson Reuters' copyright by using its legal research content to train an AI platform. The decision marks a significant moment in the ongoing debate over AI and intellectual property, as over 39 similar lawsuits progress through US courts.

Ross had argued that its use of Reuters' Westlaw headnotes, summaries of legal decisions, was transformative, meaning it repurposed the material for a different function. However, the judge rejected this defence, ruling that Ross merely repackaged the content without adding significant new value. The company's commercial intent also played a role in the ruling, as its AI system directly competed with Reuters' legal research services.

The ruling could impact future AI copyright cases, particularly those involving generative AI models trained on publicly available content. While some believe it strengthens the case for content creators, others argue its scope is limited. Legal experts caution that further court decisions will be needed to define how copyright law applies to AI training in the long term.