Google’s AI copyright problems are growing.
A group of major publishers and bestselling author Scott Turow have filed a class-action lawsuit accusing Google of using millions of copyrighted books without permission to train its Gemini AI models.
The lawsuit, filed in the U.S. District Court for the Southern District of New York, claims that Google exceeded the permissions granted through services such as Google Books, Google Play Books, and Google Scholar by repurposing copyrighted works to develop commercial AI systems.
According to the complaint, those agreements only allowed Google to display searchable snippets, provide bibliographic information, or sell ebooks. The plaintiffs argue that those agreements did not permit Google to use the material to train commercial AI models.
The plaintiffs include Hachette Book Group, Cengage Learning, Elsevier, Scott Turow, and S.C.R.I.B.E., his publishing imprint.
Allegations of copyright concealment
Beyond the alleged unauthorized use of copyrighted books, the lawsuit also alleges Google removed or altered copyright management information from the works to obscure the source of the training material.
According to the complaint, an internal Google document allegedly described the practice of using copyrighted books for AI training as “highly problematic” and warned the company it could face between $10 billion and $100 billion in potential fines if the practice became the subject of litigation.
The lawsuit is bold in its language, accusing Google of abandoning its once-famous “Don’t be evil” philosophy in its race to remain competitive in AI. It describes the company’s alleged actions as “one of the most prolific infringements of copyrighted materials in history.”
The publishers argue Gemini could pose a direct commercial threat to authors by generating content that competes with original works, including books it was allegedly trained on.
As one example, the complaint claims Gemini can produce a 100-page murder mystery novel set in a quiet seaside town in roughly 20 minutes for about $0.39, creating an inexpensive alternative for books on which the AI was allegedly trained.
The complaint specifically identifies several books that were allegedly used without permission, including The Fifth Season by N.K. Jemisin and Who Could That Be at This Hour? by Lemony Snicket.
AI legal battles continue to grow
The case represents just one phase of the rapidly expanding legal battle between copyright holders and AI developers.
Similar lawsuits have been filed against Meta, OpenAI, and Anthropic over their AI training practices. Earlier this year, Hachette and Cengage also sued Meta over allegations involving its Llama AI models, while Google remains a defendant in a separate consolidated copyright case in Northern California.
The publishing industry argues that AI companies are profiting from creative works without obtaining licenses or compensating the creators. AI developers, meanwhile, have largely argued that training models on publicly available or legally obtained content qualifies as fair use under U.S. copyright law.
A new test for fair use
The legal landscape, however, remains unsettled. Recent court rulings have largely favored AI companies. Earlier decisions in California concluded that using copyrighted works to train AI models can qualify as fair use. However, legal experts have cautioned that they do not establish binding nationwide precedent.
In 2025, Anthropic agreed to a preliminary $1.5 billion settlement to resolve claims that it used pirated books to train its Claude AI models. This highlights an important distinction that courts may have between lawfully acquired and unlawfully sourced training materials. Many authors reportedly opted out of that settlement to continue pursuing broader legal claims regarding AI training.
Because the new lawsuit against Google was filed in New York rather than California, it allows a different federal court to weigh in on how fair use applies to generative AI and whether existing licensing agreements permitted Google’s alleged conduct.
More Google coverage
What to expect next
The publishers are seeking statutory damages, a permanent injunction preventing further unauthorized use of their works, and an order requiring Google to destroy any unauthorized copies used to train Gemini. They argue that the outcome could have far-reaching consequences for the publishing industry, authors’ rights, and the future development of generative AI.
The lawsuit adds to a growing wave of AI copyright litigation that could shape how developers acquire training data and whether publishers and authors are compensated for its use.
Google had not publicly commented on the lawsuit at the time of publication.
Also read: Google AI Overviews are facing legal and regulatory tests as publishers, courts, and companies question who is responsible for AI-generated summaries.
Read the full article here
