APPLE SUED, AGAIN, FOR ALLEGEDLY USING COPYRIGHTED MATERIAL TO TRAIN ITS AI

Summary from the TLDR Design newsletter

Two neuroscientists filed the second lawsuit in just over a month against Apple, claiming the company trained its Apple Intelligence models using pirated book collections, including their copyrighted works. The lawsuit alleges Apple used Books3, a shadow library containing over 190,000 pirated books, to train its OpenELM model—the same dataset involved in other AI copyright cases. Apple joins Meta, OpenAI, and Anthropic in facing legal action over AI training practices, while tech companies typically defend their use of copyrighted material under the fair use doctrine.