Hachette Book Group, Cengage Learning, Elsevier, author Scott Turow, and his company S.C.R.I.B.E. filed a class action lawsuit against Google in New York federal court on July 14 or 15, 2026, alleging copyright infringement related to the Gemini AI project [1, 2, 3, 4, 5].

The lawsuit claims Google copied millions of copyrighted books and other works provided under limited-use agreements—such as those for Google Books and Google Play—without permission to train its Gemini AI models [1, 2, 3, 4, 5]. According to the complaint, Google stripped or removed copyright management information from these works to hide their use in AI training [2, 5]. The complaint states Gemini produces AI-generated content that directly competes with the original authors’ works and imitates their style and creative choices [1, 3, 5].

According to the lawsuit, Google was aware of the legal risks. Internal documents described using books to train AI as "highly problematic for Google" with potential fines up to $100 billion [4]. Kirk Sigmon, an intellectual property law expert, said any fair use defense might be undercut if the books were obtained unlawfully [4]. The suit also insists that Google’s use of copies intended only for search or limited display in AI training constitutes unauthorized copyright infringement [2, 4].

The plaintiffs have requested injunctions against Google and unspecified damages [1, 3]. They argue Google "reproduced millions of copyrighted works without permission, without providing any compensation to authors or publishers, and with full knowledge that its conduct violated copyright law" [5].

Similar legal battles have taken place over AI training data. In May 2026, the same publishers sued Meta for using copyrighted materials in AI development [1, 3]. In 2025, Anthropic settled a class action with authors for $1.5 billion, with a judge ruling partial fair use for their AI training [1, 3, 5]. Meta won partial fair use rulings in 2024 and 2025, though courts have not yet set a clear legal precedent on these issues [1, 2, 3, 5].

The case adds to ongoing litigation testing the limits of copyright law against AI training practices. The next procedural steps will unfold in the New York federal court following the July filing [1, 2, 3, 4, 5].