Meta accused of using pirated books for training AI with Mark Zuckerberg at the helm
Meta Platforms, led by CEO Mark Zuckerberg, is facing allegations of using pirated books to train its artificial intelligence systems. A group of authors has accused Meta of knowingly utilizing copyrighted material, as revealed in court documents. The authors claim that internal Meta documents, disclosed during the discovery phase, demonstrate the company's awareness of the pirated nature of the works. The lawsuit, filed in 2023, asserts that Meta misappropriated their books to train its large language model, Llama.
Allegations and Details
In court filings made public on Wednesday, the authors disclosed that Meta utilized the AI training dataset LibGen, which reportedly houses millions of pirated works. Furthermore, they alleged that Meta distributed this dataset through peer-to-peer torrents. Despite warnings from Meta's AI executive team, including concerns labeling LibGen as "a dataset we know to be pirated," Zuckerberg purportedly greenlit the use of LibGen.
Legal Ramifications
The authors have requested the court's approval to submit an updated complaint based on this fresh evidence. They contend that these revelations bolster their claims of copyright infringement and warrant the inclusion of new allegations, such as a computer fraud claim. This case forms part of a broader trend of lawsuits accusing tech firms of utilizing copyrighted materials without authorization to develop AI systems. Defendants often cite "fair use" as a defense in such cases.
Implications
The outcome of this legal battle could have far-reaching consequences for how AI enterprises incorporate copyrighted materials in their training procedures. At present, Meta and its CEO, Mark Zuckerberg, remain under scrutiny. The attention on Meta's practices underscores ongoing concerns about the use of copyrighted content in AI development.
