- Strike 3 Holdings and Counterlife Media accuse Meta of downloading approximately 2.400 adult films via BitTorrent
- The plaintiffs allege that the material was used to train an unannounced AI; they are seeking $359 million in damages.
- Meta denies the allegations, requests that the case be dismissed, and argues that IP addresses do not prove corporate wrongdoing.
- The company maintains that its policy prohibits explicit content and that the pattern fits with personal use.
The Menlo Park-based technology company has been singled out by a copyright claim which links it to massive downloads of adult content through its networkThe case, presented in a California federal courtreopens the debate on how data is obtained and used with artificial intelligence purposes.
The plaintiffs, Strike 3 Holdings and Counterlife Media, maintain that Since 2018, approximately 2.400 movies have been downloaded via BitTorrent from IP addresses related to Meta.They are demanding a compensation of up to 359 million, as reported by the specialized media outlet TorrentFreak.
What does the prosecution claim?
According to the lawsuit, several corporate IP addresses attributed to Meta They appear in P2P traffic tracking associated with the downloading of adult titles. The production companies claim that those downloads were part of a compilation intended to train an unannounced AI model, Linked to video generation tools such as Movie Gen or the Llama model family.
Furthermore, The plaintiffs' lawyers argue the existence of a parallel network of approximately 2.500 “hidden” IPs which would have served to conceal the origin of the trafficking. All of this supports his claim for damages, amounting to 359 million, and the request for clarification regarding the alleged use of protected material for training purposes.
Meta's defense
The company has responded with a motion to dismiss The procedure. He argues that an IP address, by itself, is not enough to attribute an infringement to the company and that there is no direct evidence that those downloads were part of a corporate plan.
Meta also points out that the chronology doesn't fit: The alleged price drops would begin in 2018, while the company's generative video projects would have taken off years later.From 2022Also, remember that their internal policies They expressly prohibit the use of explicit content to train their systems.
Another point of defense is the activity pattern. The company maintains that the volume and cadence of the files —described as downloads intermittent and scattered— They would be better suited for personal use. that with a systematic ingestion of data for AI. Along those lines, he adds that tens of thousands of employees, contractors, and visitors access the internet daily from their facilities, so It is not possible to identify specific individuals responsible..
Technical tests and questions
El The evidentiary focus revolves around BitTorrent logs and the attribution of addresses IPAlthough these tracking methods allow you to follow the trail of a file, linking a corporate IP address to a deliberate action by a company —and not to an individual connected to their network— It's a controversial point. which usually requires additional evidence.
The Plaintiffs claim to have detected a total of approximately 2.400 titles over several years, and that the traffic would be consistent with a training purpose. Meta, for its part, counters that the figure breaks down into a few dozen files per year through multiple IPs, a pattern that, in his opinion, does not fit with an AI data pipeline.
Context and interpretation from Spain and the EU
Beyond the litigation in the United States, the case resonates in Europe because of the background of copyright and data usage in model training. The increasing regulatory scrutiny—with frameworks such as the AI Regulation and the standards of digital services— points to a need for greater transparency when using protected works, especially if it is claimed that they came from unauthorized sources.
The debate comes at a time when the industry is discussing the limits of adult content in products Generative AIAlthough the lawsuit focuses on specific events, public interest also encompasses how large technology companies manage their datasets And what guarantees exist to prevent the integration of materials that infringe intellectual property?
For now, the legal battle pits two narratives against each other: one that points to an alleged use of adult content for training purposes, and another that denies any corporate plan and reduces what happened to individual activity within a broad network. The court ruling will have to assess the weight of the Technical tests, the timeline and internal policies to determine if there is sufficient basis to proceed.
I am a technology enthusiast who has turned his "geek" interests into a profession. I have spent more than 10 years of my life using cutting-edge technology and tinkering with all kinds of programs out of pure curiosity. Now I have specialized in computer technology and video games. This is because for more than 5 years I have been writing for various websites on technology and video games, creating articles that seek to give you the information you need in a language that is understandable to everyone.
If you have any questions, my knowledge ranges from everything related to the Windows operating system as well as Android for mobile phones. And my commitment is to you, I am always willing to spend a few minutes and help you resolve any questions you may have in this internet world.