Authors file lawsuits against Meta

Several hundred authors have filed lawsuits against Meta, accusing the company of using their works for training artificial intelligence models without permission. Nearly 300 new complaints from authors were added to existing procedures in the United States and around the world on October 19, 2026, according to reports.

Legal actions supervised by two firms

The legal actions are being supervised by two law firms, Freedman Normand Friedland LLP and Stris & Maher LLP, within the district court for the northern district of California. The authors are demanding up to $150,000 in damages and interest from Meta, which had not responded to the article at the time of publication.

French book organizations sue Meta

In France, several book organizations, including the National Syndicate of Publishing, the Society of Writers and Composers, and the National Syndicate of Authors and Composers, have sued Meta since March 2025 for "infringement" and "economic parasitism," also accusing the group of using protected works without authorization to train its AI Llama.

Confidence in legal outcomes

Vincent Montagne, president of the SNE, expressed confidence in the progress of the procedure, stating that "we have hundreds of thousands of evidence, Meta will be condemned, there is absolutely no doubt."

Individual approach in legal strategy

The strategy differs from other cases already brought before American courts, as the lawyers and their clients this time preferred the individual approach rather than a class action. The law firms state that more than a thousand authors in total have entrusted them with their files.

Previous court rejection of claims

In another case, targeting Meta, the court rejected the authors' requests, considering that their arguments were not strong enough to prove the existence of a financial prejudice resulting from the unauthorized use of their works for training purposes.

Rise in AI training lawsuits

In the United States, procedures have multiplied in recent years around the training conditions of large language models, directly linked to generative artificial intelligences. To make these more effective, AI developers have resorted to huge volumes of data to "train" the models.

Accusations of copyright violations

This has led to accusations that several generative AI developers, including Anthropic, Nvidia, OpenAI, and Meta, have used pirated book reservoirs still covered by copyright. The use of this copyrighted content has been extensively documented, and AI developers have developed another strategy to respond to the accusations, citing fair use provisions of American law.

Settlement with Anthropic

In a separate case, Anthropic reached a $1.5 billion settlement with holders of rights of hundreds of thousands of books, which has now opened disputes over the distribution of rights among authors, publishers, and literary agencies. The French publishing industry aims to use this case to encourage technology companies to establish partnerships and acquire legal exploitation licenses.

Focus on legal rights during AI inference

The SNE has also highlighted non-compliance with the right of opposition by AI actors during the so-called inference phases.