Authors' lucky break in court may help class action over Meta torrenting

2026-03-30 · Source: AI - Ars Technica · Field: Legal & Regulatory — Intellectual Property & Patents, Litigation & Dispute Resolution, Compliance & Risk Management · Depth: Intermediate, medium

Summary

Meta faces ongoing legal challenges over its alleged torrenting of 80 terabytes of pirated books to train AI models, with authors pursuing both direct and contributory copyright infringement claims. A recent Supreme Court ruling, *Cox v. Sony*, found ISPs not liable for piracy on their networks, which Meta hopes will shield it from liability. However, a judge recently allowed a contributory infringement claim to be added to the *Kadrey v. Meta* class action, making it easier for authors to prove Meta facilitated copyright transfers without needing to show full work seeding. This decision, despite the judge's criticism of the authors' lawyers for delays, was influenced by Meta's own request to align case schedules, preventing prejudice. Meta's defense hinges on arguing it did not "affirmatively induce" infringement, while plaintiffs contend Meta's knowledge of BitTorrent's upload-dependent design constitutes certain knowledge of infringement.

Key takeaway

For legal teams and AI developers navigating copyright claims related to training data, your understanding of secondary liability standards is critical. The *Cox v. Sony* ruling, while seemingly beneficial for service providers, may not fully shield entities actively involved in data acquisition via torrenting. You should meticulously document data sourcing and internal discussions to preempt claims of "affirmative inducement" or "certain knowledge" of infringement, as these will be key battlegrounds in future litigation.

Key insights

The *Cox v. Sony* Supreme Court ruling on ISP liability for piracy is now central to Meta's AI training data copyright defense.

Principles

• Contributory infringement is easier to prove than direct distribution.
• Knowledge of a system's design implies knowledge of its consequences.

In practice

• Aligning case schedules can inadvertently benefit opposing parties.
• Late claim additions can be approved to protect class member interests.

Topics

• Copyright Infringement
• AI Training Data
• BitTorrent Piracy
• Contributory Infringement
• Supreme Court Precedent

Best for: CTO, VP of Engineering/Data, Director of AI/ML, Legal Professional, Tech Journalist, Policy Maker

Related on AIssential

• Judge Rules Blacked.com Can Sue Meta for Scraping Its Porn — Copyright infringement occurs upon unauthorized downloading and distribution, regardless of AI training intent.
• Elsevier v. Meta: Not just another “AI trained on copyrighted works” lawsuit. It is drafted as a story of deliberate corporate piracy, executive authorisation, concealment, and market substitution. — The Elsevier-Meta lawsuit alleges deliberate corporate piracy, not just AI training on copyrighted works.
• Cognella v. Anthropic: Anthropic allegedly acquired pirated books from shadow libraries, torrented them, redistributed them through peer-to-peer networks, stripped copyright management information... — AI copyright litigation is shifting from abstract "training" debates to concrete allegations of piracy and illicit content acquisition.
• How AI Companies Can Pay Fair Rates for the Content They Need — The core conflict is AI companies using content freely versus creators demanding fair compensation.
• Microsoft trained its MAI models on unlicensed web data despite promising "enterprise grade, clean and commercially licensed data" — Microsoft's MAI models used unlicensed web data, challenging its "clean data" claims amidst ongoing fair use legal disputes.
• Penguin Random House Sues OpenAI — AI memorization and reproduction of copyrighted content, including text and visuals, faces increasing legal challenges.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by AI - Ars Technica.