On September 5, 2025, a complaint was filed against Apple Inc. before the United States District Court for the Northern District of California, San Francisco Division, by authors Grady Hendrix and Jennifer Roberson. This marks Apple’s entry into the expanding wave of copyright litigation over the use of protected works to train artificial intelligence systems. While OpenAI, Meta, and Anthropic are already defendants in similar actions, Apple had until now largely avoided this trend.
The plaintiffs allege that Apple copied entire literary works without authorization in order to build the datasets underlying its “Apple Intelligence” models. The complaint asserts direct copyright infringement under 17 U.S.C. § 501 and seeks damages, disgorgement of profits, and a permanent injunction. More strikingly, the plaintiffs invoke 17 U.S.C. § 503(b) to request the destruction of Apple’s AI models trained on infringing material. The case has been filed as a putative class action on behalf of all U.S. copyright owners with registered works allegedly used in Apple’s training process.
This litigation raises fundamental legal questions at the heart of AI regulation in the United States: whether copying works from so-called “shadow libraries” constitutes infringement per se, whether the fair use doctrine can apply in the context of large-scale machine learning, and how to assess datasets described as “publicly available” or “open-sourced” when they incorporate copyrighted works. The outcome could have profound consequences for the future licensing market in AI training.
