On 10/7, three publishers—Hachette Book Group, Cengage Learning, and Elsevier—along with American author Scott Turow and S.C.R.I.B.E. Inc., Turow's publishing and rights management company, filed a lawsuit against Google in the U.S. District Court for the Southern District of New York. The complaint was published by the Association of American Publishers (AAP) on the same day. According to the AAP, the plaintiffs also requested the court certify the case as a class action, representing authors and publishers with similar interests.

The plaintiffs allege that Google unlawfully copied millions of books and academic articles to develop its Gemini models. They describe this as one of the "most massive copyright infringements in history." The U.S. District Court for the Southern District of New York has not yet ruled on the merits of the case. Google has not yet responded to the allegations.

According to the complaint, Google "secretly copied" works provided to its Google Books, Google Play Books, and Google Scholar services. Authors and publishers had only granted these services limited rights to exploit the materials, specifically for purposes such as displaying excerpts for search results or selling books. However, the plaintiffs claim that Google used this data to train its commercial artificial intelligence (AI) products, as reported by Guardian.

The Google Gemini logo on a phone application. Photo: Bao Lam

The filing further alleges that Google collected text from large-scale web scans, including pirated book repositories and paid news sources. The plaintiffs also accuse the company of "removing or altering copyright management information" to conceal the origins of materials used for Gemini.

The plaintiffs contend that Google neither sought permission nor paid for the works incorporated into its training process. The complaint cites documents described by the plaintiffs as internal company materials, which show that the use of copyrighted books provided by publishers via Google Play Books was considered "particularly concerning for Google."

To demonstrate Gemini's use of book content, the plaintiffs cited a chatbot's response regarding N. K. Jemisin's 2015 fantasy novel, The Fifth Season. The system stated its internal database contained text from the work, enabling it to reproduce plot, characters, terminology, and quotes.

Other book titles cited by the publishers as having their copyrights infringed include Who Could That Be at This Hour? (2012) by Lemony Snicket, and Innocent (2010) by Scott Turow, along with numerous other works, textbooks, and scientific articles. However, these details represent an incomplete list of evidence provided by the plaintiffs. Google has not yet commented on the matter, according to Guardian.

The plaintiffs argue that Google's actions could reduce book sales revenue and impact the market for licensing content to AI. The complaint asserts that authors and publishers lose potential income, bargaining power over usage terms, and opportunities to demand attribution when Google fails to secure agreements before using their works.

The complaint provides an example: Gemini could write a "100-page murder mystery set by the sea in a mysterious town" to replace a copyrighted original story it was trained on. "It can generate a series in just 20 minutes for USD 0,39. No publisher or writer can compete with that speed and scale," the plaintiffs stated.

This lawsuit extends the legal disputes between AI and book copyright, according to Guardian. To date, authors and publishers have filed numerous lawsuits against AI companies like Google, OpenAI, Anthropic, and Meta, alleging unauthorized copyright infringement for source data.

Since 2023, American actress Sarah Silverman, author of the memoir The Bedwetter: Stories of Courage, Redemption and Pee, and other writers have sued Meta AI for using their works in its source data. In 6/2025, the court ruled in favor of Meta, finding that the plaintiffs had not provided sufficient evidence of market damages.

In 9/2025, Anthropic agreed to establish a USD 1,5 billion fund to settle a class-action lawsuit brought by authors concerning books allegedly sourced from pirated content. The agreement did not include an admission of liability by the company.

Prior to the current lawsuit, on 15/1, Hachette and Cengage publishers had sought to join an existing lawsuit between authors and illustrators against Google over copyrighted images in their works. According to Reuters, the two publishing companies argued they could provide evidence of damages for the plaintiffs. However, Google opposed the request. Following that denial, the publishers decided to pursue their own action, Guardian reported.

The plaintiffs are seeking statutory damages or actual damages and related profits from Google. They also demand that the company cease its alleged infringing activities, disclose information about its training data, and destroy identified infringing copies under court supervision.

Since the AAP publicly released the complaint, Google has not responded to requests for comment on the lawsuit. All arguments regarding how Google collected, copied, or used the works have been presented by the plaintiffs.

The Association of American Publishers was established in 1970, following the merger of the American Book Publishers Council and the American Educational Publishers Institute. The non-profit organization is currently led by copyright law expert Maria A. Pallante. It represents the U.S. publishing industry on legal and policy matters, with a focus on copyright. The AAP currently has over 110 member organizations, including major publishing corporations, independent entities, university presses, and professional associations.

Thao Uyen (according to Guardian, AAP, Reuters)