Surveys, opinion polls, and media analysis

Authors Accuse Meta of Using Pirated LibGen Dataset to Train Llama AI, Zuckerberg Approved

Jan 10, 2025, 10:41 AM

A group of authors, including Ta-Nehisi Coates and Sarah Silverman, have accused Meta Platforms Inc. of using pirated versions of copyrighted books to train its artificial intelligence systems. According to court filings in a California federal court, Meta's CEO Mark Zuckerberg approved the use of the LibGen dataset, despite internal concerns that it contained pirated materials. The authors allege that Meta not only downloaded these materials but also stripped copyright information from them to train its Llama AI models. The lawsuit claims that Meta's actions constituted copyright infringement, and the company attempted to conceal its use of the pirated dataset. Meta has argued that its use of the materials falls under the fair use doctrine, a claim that has been contested in previous similar lawsuits against the company.