Five major publishers and a best-selling author sued Meta on May 5, 2026, claiming the company trained its Llama AI platform on millions of pirated books and articles without authorization or compensation. The complaint alleges that Meta CEO Mark Zuckerberg personally authorized the infringement. This filing escalates the already-contested legal position on training-data IP: comes after federal court rulings (June 2025) on fair use and ongoing publisher litigation across the AI vendor ecosystem, signaling that IP provenance risk is moving from abstract compliance concern to concrete litigation exposure with potential damages and operational constraints on training data sourcing.
356 other tracked developments touching the same claims.