Thomson Reuters has won The first grave case AI Copyright in the United States.
In 2020, the conglomerate of media and technology submitted an unprecedented trial on copyrights to AI against the legal startup AI Ross Intelligence. In the complaint, Thomson Reuters won the company AI recreated materials from its research company Westlaw. Today, the judge ruled in favor of Thomson Reuters, stating that the copyright to the company was actually violated by Ross Intelligence.
“None of the possible defense of Ross holds water. I reject them all, “wrote the District Court of Judge Delaware Stephanos Bibas, in a summary judgment.
Thomson Reuters and Ross Intelligence did not immediately respond to requests for comment.
The AI generative boom led to a number of additional legal fights for how AI companies can apply copyright protected materials, because many main AI tools were developed by training on copyright protected works, including books, films, visual graphics and visual graphics and websites. Currently, there are currently several dozen lawsuits, which are currently winding in the American judicial system, as well as international challenges in China, Canada, Great Britain and other countries.
In particular, Judge Bibas ruled in favor of Thomson Reuters in terms of permitted apply. The doctrine of allowed apply is a key element of how AI companies try to defend themselves against claims that they illegally used copyright protected materials. The idea underlying the allowed apply is that sometimes it is legally permissible to apply works protected by copyright without permission – for example, creating works of parody or in non -commercial research or production of messages. When determining whether the permitted apply is required, the courts apply a four -factor test, looking at the cause of work, the nature of the work (whether poetry, fact literature, private letters, etc. As the apply affects the market value of the original. Thomson Reuters won two of the four factors, but the Bibas He described the fourth as the most critical, and ruled that Ross “was to compete with Westlaw, developing a market substitute.”
The spokesman for Thomson Reuters Jeffrey McCoy applauded the decision in a statement sent to Wired. “We are glad that the court issued a summary judgment in our favor and came to the conclusion that the editorial content of Westlawa created and maintained by our lawyer editors are protected by copyright and cannot be used without our consent,” he wrote. “Copying our content was not” allowed “.
Even before this ruling, Ross Intelligence had already felt the influence of a court battle: Startup close In 2021, citing the expense of court disputes. However, many AI companies are still throwing him in court, like Opeli and Google, are financially equipped with extended legal fights.
Despite this, this ruling is a blow to AI, according to the professor of the University of Cornell of Digital and Internet James Grimmelmann: “If this decision is observed elsewhere, it is really harmful to AI generative companies.” Grimmelmann believes that the Bibas judgment suggests that many case -law cite AI generative companies to argue that the permitted use is “irrelevant”.
Chris Mammen, a partner at Wombble Bond Dickinson, who focuses on intellectual property law, recognizes that this would complicate the arguments of the permitted use by AI, although they may vary depending on the plaintiff. “It puts a finger on a scale in the direction of maintaining that there is no allowed apply,” he says.
Update 2/11/25 5:09 ET: This story has been updated to contain an additional comment Thomson Reuters.