A group of prominent YouTube creators has launched a class-action lawsuit against Snap Inc., alleging the tech giant illegally scraped their videos to train artificial intelligence models. The plaintiffs, representing channels with a combined 6.2 million subscribers, claim their intellectual property was utilized without consent to power AI features, including the app’s “Imagine Lens.”
Legal Battle Over Unauthorized AI Training
Filed this past Friday in the U.S. District Court for the Central District of California, the lawsuit specifically targets Snap’s reliance on the HD-VILA-100M dataset. The creators argue that Snap circumvented YouTube’s terms of service and technological barriers to repurpose academic and research-only datasets for commercial gain.
The legal action is spearheaded by the team behind h3h3, alongside the golf-focused channels MrShortGame Golf and Golfholics. The plaintiffs are seeking statutory damages and a permanent injunction to prevent further unauthorized use of their content.
A Growing Trend in AI Litigation
This filing marks another escalation in the escalating conflict between content creators and AI developers. The same group of plaintiffs has previously initiated similar legal proceedings against industry heavyweights, including Nvidia, Meta, and ByteDance.
According to the Copyright Alliance, over 70 copyright infringement cases have been filed against AI companies to date. These disputes involve a diverse range of stakeholders, from newspapers and book authors to artists and user-generated content platforms.
The Uncertain Future of AI Copyright
The judicial landscape remains fragmented. While some courts have ruled in favor of tech corporations—such as in the case involving Meta and a group of authors—other disputes have ended in settlements, like the agreement between Anthropic and its plaintiffs. Many of these high-stakes cases remain in active litigation, leaving the legal status of AI training data in a state of flux.
Snap has officially declined to comment on the pending litigation.