A group of YouTubers suing tech giants for scraping their videos without permission has added Snap to their list of defendants. The plaintiffs, creators behind three YouTube channels with roughly 6.2 million subscribers, allege Snap trained its AI systems on their content for features such as the “Imagine Lens,” which edits images using text prompts.
The plaintiffs previously filed similar lawsuits against Nvidia, Meta, and ByteDance. The new proposed class-action, filed in the U.S. District Court for the Central District of California, cites Snap’s use of the HD-VILA-100M video-language dataset, originally intended for academic and research purposes, claiming the company bypassed YouTube’s technological and licensing restrictions for commercial gain. The suit seeks statutory damages and a permanent injunction against further use of the videos.
The case is led by the h3h3 YouTube channel and two smaller golfing channels, MrShortGame Golf and Golfholics, and underscores the increasing legal scrutiny faced by AI companies over the use of copyrighted content. This lawsuit adds to a growing wave of similar challenges targeting technology firms accused of training AI models on copyrighted material without permission. One of the recent examples is Adobe, which is being sued for allegedly using pirated books to train its SlimLM AI model.
At the same time, some companies are taking steps to ensure responsible AI development. For example, Universal Music Group is partnering with Nvidia to create AI tools for music discovery and creation while guaranteeing proper artist attribution and copyright protection, demonstrating how the industry is attempting to balance innovation with legal and ethical standards.
