In an effort to gather materials for LLM training, Meta used BitTorrent to download pirated books from Anna's Archive and other libraries. According to several authors, Meta facilitated the infringement by “seeding” these torrents. This week, the court gave the authors permission to add these allegations to their lawsuit, despite openly berating their lawyers for “stupid excuses” and “attacking Meta.”
Over the past two years, rights holders of all kinds have filed lawsuits against companies that develop Artificial Intelligence models.
Most of these cases allege that AI developers are using copyrighted works to train LLMs without obtaining permission.
Meta is among a long list of companies that have been sued for this activity. This includes a class action lawsuit filed by authors including Richard Kadrey, Sarah Silverman, and Christopher Golden, which accused Meta of using libraries of pirated books as educational materials.
The Court initially rejected the plaintiffs' claims.
Last summer, Meta scored a significant victory in this case, as the court concluded that using pirated books to teach Llama LLM was “fair use,” based on the arguments presented in that case. However, it was a bittersweet victory, as Meta remained liable for downloading and sharing the books via BitTorrent.
By downloading books from libraries like Anna's Archive, Meta used BitTorrent. So in addition to downloading the content, the app also uploaded the downloaded data and shared it with others. According to the authors, this means that Meta was engaging in widespread and direct copyright infringement.
Last December, the authors, through their lawyers, requested permission to file a fourth amended complaint. Specifically, they wanted to add that seeding is a copyright infringement, as Meta made it easier for others to download.
Although the press releases will range from very select to rare, I said I'd pass...because sometimes the editors hide.
