In a dramatic twist in the ongoing copyright lawsuit, Meta has defended its practice of torrenting a dataset of pirated books for AI training by arguing that simply downloading files using a widely-used protocol does not equate to illegal seeding. In its court filing, Meta stated it “took precautions not to ‘seed’ any downloaded files,” suggesting that absent proof of sharing the files further, the act of torrenting does not breach copyright laws. This defense comes amid allegations from well-known authors—such as Richard Kadrey, Sarah Silverman, and Ta-Nehisi Coates—who accuse Meta of orchestrating one of the largest data piracy campaigns in history, intentionally bypassing legal licensing systems. The authors contend that Meta’s torrenting not only facilitated unauthorized distribution of millions of copyrighted works but also deprived them of revenue, a claim underscored by their statement, "Meta’s decision to bypass lawful acquisition methods and become a knowing participant in an illegal peer-to-peer piracy network provides the 'extra element' and is 'qualitatively different'." As the case unfolds, both sides present compelling testimony involving modified torrent settings and attempts to conceal seeding activity, leaving the interpretation of torrenting terminology—and its legal ramifications—in the hands of the court.

Key Points:

• Torrenting Defense: Meta claims that torrenting, a method for downloading large files, is not inherently illegal, emphasizing that there is no proof they seeded (shared) the material beyond the download.
• Court Filing Details: Meta's filing states they “took precautions not to ‘seed’ any downloaded files,” arguing that without evidence of seeding, distributing the pirated books cannot be legally substantiated.
• Accusations by Authors: Prominent figures including Richard Kadrey, Sarah Silverman, and Ta-Nehisi Coates accuse Meta of conducting a massive data piracy operation, asserting that the company bypassed lawful licensing methods.
• Legal Implications: The authors argue that if Meta had legally bought or borrowed the books, it would be a simple copyright infringement, but by engaging in torrenting from public repositories like LibGen and Z-Library, Meta committed an independent violation under California’s Computer Data Access and Fraud Act.
• Testimonies and Evidence: Testimony from a Meta executive and internal messages from a researcher indicate modifications to torrenting settings to minimize seeding, although these actions may still not absolve the company entirely, according to critics.
• Ongoing Legal Battle: Meta is now aiming to dismiss the CDAFA claim, insisting that its torrenting activity was merely a means to access data from widely available public sources, while the case continues to expose the complex legal issues surrounding AI training and copyright law.

This legal saga not only highlights the murky waters of torrenting technology but also underscores the broader debates over data rights in the age of artificial intelligence.
Link to Article