Thomson Reuters wins lawsuit against AI company over copyright infringement

Press agency Thomson Reuters has won a lawsuit against Ross Intelligence, an AI company that used the contents of Reuters’ legal research database Westlaw without permission to benefit its own legal search engine.

Thomson Reuters owns Westlaw, which is a major legal research platform that provides a variety of legal sources, including case law, court decisions, statutes, legal articles, and editorial summaries written by editors, also called headnotes. It requires a paid subscription, which costs hundreds of dollars per month. Westlaw is therefore mainly used by law firms and legal institutions.

Ross Intelligence is a legal AI startup and a competitor that has developed a legal AI search engine. The company tried to license Westlaw’s content, but Reuters refused this request. Instead, Ross decided to make a deal with LegalEase to get training data in the form of so-called ‘Bulk Memos’. Bulk Memos are compilations of legal questions and answers based on Westlaw’s headnotes.

According to court documents, Ross Intelligence built its competing product using Bulk Memos, which in turn were built from Westlaw headnotes. When Thomson Reuters found out, it sued Ross for copyright infringement.

The United States District Court for the District of Delaware has decided in favor of the press agency. The judge ruled that Westlaw’s headnotes are subjected to copyright protection. Ross Intelligence copied these headnotes to use as training data for its legal AI search engine.

In addition, the court ruled that fair use doesn’t apply in this case, because Ross Intelligence’s use is commercial and not transformative: it doesn’t have a different character compared to Westlaw, because Ross’ legal research tool was built to compete with Thomson Reuters’ tool.

“We are pleased that the court granted summary judgment in our favor and concluded that Westlaw’s editorial content created and maintained by our attorney editors, is protected by copyright and cannot be used without our consent. The copying of our content was not ‘fair use’,” Thomson Reuters spokesperson Jeff McCoy told The Verge in a statement.

This ruling sets a precedent for AI companies for prohibiting the use of copyrighted materials for training purposes without consent, and whether they can claim ‘fair use’ as a defense for using copyrighted materials.

A similar lawsuit has been brought up against OpenAI and Microsoft by The New York Times, which is accusing the companies to have trained their AI models by copying and using millions of the news outlet’s articles, and thus becoming the publication’s direct competitor.