A group of leading publishers and authors has taken Google to court, alleging that the tech giant used their copyrighted books to train its Gemini artificial‑intelligence system without permission. The lawsuit, filed in the U.S. District Court for the Southern District of New York, names Hachette, Cengage, Elsevier, author Scott Turow and the author‑rights collective S.C.R.I.B.E. as plaintiffs.
According to the complaint, Google copied entire works from programs such as Google Books and Google Play, then fed them into Gemini’s training pipeline. The plaintiffs argue that the company stripped or altered copyright metadata to conceal the fact that it was building AI models on material it never authorized to use.
"Google illegally copied works from all these scope‑limited programs for AI training, knowing it lacked authorization to do so," the filing says. The suit also points to an internal Google document that reportedly warned senior staff that using copyrighted books for AI could expose the company to fines ranging from $10 billion to $100 billion.
Google’s relationship with the publishing industry dates back to the early 2000s, when it launched Google Books to make snippets of books searchable while keeping full texts behind paywalls. Publishers have long allowed Google to index limited portions of their titles in exchange for increased discoverability. The plaintiffs maintain that the company crossed a line by harvesting entire copies for machine‑learning purposes, a use not covered by the original agreements.
The case arrives amid a broader legal battle over AI training data. California courts have issued two early rulings that favored AI companies, deeming large‑scale use of copyrighted works for training as "fair use." Those decisions, however, have not settled the issue nationwide. In a separate matter, Anthropic, another AI firm, was ordered to pay a $1.5 billion penalty for allegedly pirating the works it used to train its models, the largest copyright settlement in U.S. history.
While the Anthropic verdict underscores the potential financial risk for tech firms, the California rulings suggest that courts may still be inclined to protect AI development under the existing fair‑use doctrine. The New York filing gives judges a fresh venue to weigh the competing interests of innovation and copyright protection.
Google has not responded to requests for comment. The company’s silence leaves the public uncertain about how it will defend its practices or whether it will seek a settlement.
Legal experts note that the outcome could reverberate across the tech industry. A ruling against Google might force AI developers to renegotiate data‑licensing agreements with content creators, potentially reshaping how training datasets are assembled.
For now, the lawsuit adds another high‑profile chapter to the ongoing clash between the publishing world and the rapidly expanding AI sector, a clash that could define the balance between intellectual‑property rights and technological progress for years to come.
Cet article a été rédigé avec l'assistance de l'IA.
News Factory APP - actualités agentiques pour booster votre SEO et AEO.