The US-based publishers Hachette, Cengage, and Elsevier have sued Google over the use of their content in training the Gemini AI model. Their claim: Google incorporated copyrighted works into its training data without authorization from authors or publishers. This escalates the conflict between the AI industry and publishing significantly.
The essentials
- Three major publishers (Hachette, Cengage, Elsevier) file suit against Google
- Allegation: Unauthorized use of copyrighted works in Gemini training
- This opens a new legal front – similar lawsuits already exist against OpenAI and other AI providers
- The case could set precedent for the entire AI industry
The context: Why now?
This lawsuit is not isolated. Authors and publishers previously sued OpenAI and Meta with comparable claims. Now the focus shifts to Google and Gemini. The core dispute remains: Can AI providers simply scrape books, articles, and other texts to train models without paying rights holders or seeking permission?
The three plaintiffs rank among the world's most influential publishing groups. Hachette is one of the "Big Five" English-language publishers, Cengage leads the education market, and Elsevier dominates scientific publishing. That these major players are taking Google to court signals the issue is existential for the industry.
What's at stake?
The lawsuit touches fundamental questions:
| Question | Implication |
|---|---|
| Is scraping without license permitted? | Determines whether AI training can be free |
| Who profits from AI models? | Publishers demand a share of value |
| How is "fair use" defined? | Central to US jurisprudence |
A victory for publishers could mean Google and other AI providers must purchase licenses—or retrain models with different data. This would reshape the entire industry's business model.
What it means for German companies
This development matters for German enterprises. First: If US courts rule in publishers' favor, European courts may follow—the EU AI Act doesn't yet address such questions precisely. Second: German publishers must decide how aggressively to defend their rights. So far, German lawsuits against AI providers are rarer than in the Anglo-American sphere.
The open question: Will Google, OpenAI, and others ultimately strike licensing deals with publishers—or will litigation drag on while AI models are already in use?
Sources
Editorially owned by Ideal Syka. Sources and method: Newsroom & method. Tips and corrections: ai@i6eal.de.
