A recent lawsuit claims that Mark Zuckerberg and senior Meta executives approved the use of pirated content from the Library Genesis dataset to train the company's AI models, despite internal warnings about its legality. The plaintiffs, a group of authors including Sarah Silverman, have accused Meta of misusing their copyrighted works without authorization. The allegations assert that Zuckerberg knowingly disregarded concerns from within Meta's AI team about the legal ramifications of using the LibGen dataset, which hosts over 33 million books and operates outside copyright law. The lawsuit seeks damages and an injunction to prevent further use of the stolen works. Internal communications highlighted a reluctance among Meta engineers concerning the use of pirated materials. This case occurs amidst broader scrutiny of AI companies regarding copyright infringement. Meta has faced previous legal issues over similar practices, prompting substantial concern from the creative community regarding the ethical implications of AI training methods.

Source 🔗