Youngblood v. Nvidia Corporation

The Complaint alleges that Plaintiff Devin Youngblood, an Alabama resident, owns and operates "The Youngblood's Podcast" YouTube channel, which publishes weekly audio and video episodes discussing marriage, parenting, family relationships, and mental health based on personal experiences. The Complaint contends that at least one video from Plaintiff's channel appears in the HowTo100M dataset, which Defendant used to train its Cosmos model.

The Complaint alleges that YouTube employs technological protective measures to control access to underlying audiovisual files and prevent bulk extraction by automated systems, making content available primarily through controlled streaming pathways and authorised features. YouTube's terms of service expressly prohibit scraping, unauthorised downloading, bulk extraction, and other forms of data mining except through permitted product features or licensed APIs.

The Complaint contends that Nvidia, without obtaining licences or permissions from Plaintiff and class members, used circumvention tools and services to defeat YouTube's technological measures at scale to acquire file-level copies of creators' works for training its AI models. The Complaint alleges this information derives from Nvidia employees recently uncovered through media sources. The Complaint asserts that obtaining file-level copies at the scale required for modern AI video-model training necessarily requires bypassing YouTube's controls.

The Complaint brings one cause of action for violations of the DMCA's anti-circumvention provisions under 17 U.S.C. § 1201(a), which prohibits circumventing technological measures that effectively control access to copyrighted works and trafficking in circumvention technologies and services. Plaintiff seeks injunctive relief, statutory damages, and other remedies authorised by 17 U.S.C. § 1203.