Meta Platforms Inc., the parent company of Facebook, Instagram, and Threads, is facing a massive $359 million lawsuit after being accused of illegally downloading more than 2,000 adult videos allegedly for “personal use.” The complaint, filed by adult entertainment studios Strike 3 Holdings and Counterlife Media, accuses Meta of large-scale copyright infringement tied to AI data collection.
Allegations Against Meta
According to the lawsuit, Meta’s corporate IP addresses were used to download and seed approximately 2,396 copyrighted adult film titles. The plaintiffs claim that the downloads were not isolated incidents, but part of a deliberate effort to collect “diverse and explicit” visual data to enhance Meta’s generative AI systems — including projects linked to video synthesis and multimodal learning.
Investigators reportedly traced 47 primary IP addresses associated with Meta’s headquarters and a further network of over 2,500 addresses allegedly used to torrent or seed copyrighted material over several years. Strike 3 Holdings argued that such a pattern indicated “systematic and willful” infringement rather than accidental or unauthorized use by individuals.
The lawsuit asserts that the content, which includes explicit video material, was used to train Meta’s video and image-based AI models, potentially giving the company a competitive advantage in developing realistic AI-generated visuals.
Meta’s Response
Meta has strongly denied the allegations, calling the lawsuit “baseless and speculative.” In its court filing, the company said the alleged downloads were not conducted by Meta engineers or AI researchers but were instead “personal use” activity by individuals who may have accessed its corporate network.
The average download rate—roughly 22 videos annually over a seven-year period—is incongruous with any type of AI data collection operation, according to Meta’s legal team. Citing adherence to both internal ethics standards and platform restrictions, the company also emphasized its internal procedures, which expressly forbid the inclusion of adult content in model training datasets.
Additionally, Meta pointed out that it is irrational to assign all network activity to corporate purpose because its extensive worldwide network infrastructure includes visitors, temporary workers, and third-party contractors who would have been accountable for the downloads.
The case represents one of the most high-profile intersections between copyright law and AI development to date. If proven, the allegations could redefine how AI companies source and license training data particularly concerning explicit or sensitive material.
Experts believe the lawsuit could open the door to stricter oversight on how tech firms collect and use visual data for machine learning, especially in cases where copyrighted material is involved. Legal analysts also note that this could pressure regulators to introduce clearer rules on “data provenance” in AI training.
The incident also raises reputational questions for Meta. As a leading advocate of “responsible AI” development, the notion that adult films were used even indirectly in AI research could damage the company’s credibility and public trust.
Meanwhile, Strike 3 Holdings, the main plaintiff, has faced criticism in the past for aggressive copyright enforcement tactics, often referred to as “copyright trolling.” Some experts suggest the case may test whether Meta’s global visibility makes it a prime target for high-value settlements.
Preliminary review is presently underway for the lawsuit, which was filed in the U.S. District Court for the Northern District of California. Since no specific connection has been found between the purported downloads and any Meta employee or AI project, Meta has moved to dismiss the complaint.
The case may move forward to discovery if the motion to dismiss is rejected, which may compel Meta to reveal internal logs, employee activity logs, and records regarding their AI data procedures. Any discovery phase, according to industry observers, might have a significant impact on how big AI companies defend their data sourcing practices.
