Copyright Cases Roundup

Updates In Creative Content Lawsuits Challenging Generative AI Programs

On November 22, the U.S. District Court for the Southern District of New York denied OpenAI generative programs’ defense motion to compel that included the production of The New York Times plaintiff’s “use of nonparties’ generative artificial intelligence (‘Gen AI’) tools” and “the Times’s position regarding Gen AI (e.g., positions expressed outside of litigation, knowledge about the training of third-party Gen AI tools using the Time’s works).” The New York Times Co. v. Microsoft Corp., 1:23-cv-11195 (S.D.N.Y.). In the ruling, Southern District Magistrate Judge Ona T. Wang noted the litigation “is not a referendum on the benefits of Gen AI, on Plaintiff’s business practices, or about whether any of Plaintiff’s employees use Gen AI at work. The broad scope of document production sought here is simply not relevant to Defendant’s purported fair use defense. For example, if a copyright holder sued a video game manufacturer for copyright infringement, the copyright holder might be required to produce documents relating to their interactions with that video game manufacturer, but the video game manufacturer would not be entitled to wide-ranging discovery concerning the copyright holder’s employees’ gaming history, statements about video games generally, or even their licensing of different content to other video game manufacturers.”
On November 25, in Kadrey v. Meta Platforms Inc., 23-cv-03417 (N.D. Calif.), one of several class-action lawsuits book authors have filed alleging generative AI defendants are liable for copyright infringement for the unlicensed feeding of the plaintiffs’ works into large language models, Magistrate Thomas S. Hixon of the U.S. District Court for the Northern District of California in part granted and in part denied plaintiffs’ discovery motions to compel. Among these, on the motion to compel “Communications Concerning any licensing copyrighted works that were used to train the Meta Language Models,” the federal magistrate ordered Meta “to produce responsive documents regardless of whether the communications successfully resulted in a license ….” Magistrate Hixon also ordered Meta to produce: “All Documents and Communications sufficient to show Your actual or projected income from the sale or licensing of the Meta Language Models” and “All Documents and Communications Concerning any income statement, balance sheet, or statement of cash flows, Concerning any of the Meta Language Models” — though Meta claims it has already handed over the last two document categories. But the magistrate denied the Kadrey plaintiffs’ motion to compel “Documents and Communications sufficient to show each instance within the last three years where You have licensed copyrighted works for Meta's commercial use.” “The Court agrees with Meta that this [Request for Production] is unreasonably overbroad,” Magistrate Hixon explained, “because it seeks information concerning each instance in which Meta licensed a copyrighted work for Meta’s commercial use, regardless of whether the commercial use had anything to do with AI or any issue that is relevant to this case. This would include, for example, licensing a song to use in an advertisement.”