Nvidia Sued for Scraping YouTube (www.404media.co)

YouTube creator David Millette has filed a class action lawsuit against Nvidia, alleging the company scraped videos from YouTube and other platforms en masse to train a commercial “video model” without creators’ permission. The suit — prompted by a 404 Media investigation that found large-scale scraping of user content — accuses Nvidia of unjust enrichment, violating California’s Unfair Competition Law, and even breaching federal labor laws in the process of building its AI systems. Millette claims Nvidia profited from his and others’ work while denying creators compensation or consent. The case is significant because it targets a core industry practice: large-scale web scraping to assemble training datasets for generative and video-understanding models. If the court recognizes these claims, it could reshape legal obligations around dataset provenance, platform terms of service, licensing, and creator remuneration, and force greater transparency about what content companies ingest and how it’s used. Technical implications include potential limits on training data sources, stricter auditing of dataset composition, and incentives to develop licensed or synthetic datasets. The lawsuit could set precedent on whether scraping publicly available online video for commercial model training constitutes unfair competition or unjust enrichment, with wide-reaching effects on how AI/ML teams build and deploy models that rely on web-harvested media.