Meta Faces Copyright Lawsuit Over Allegations of Pirated AI Training Data

Plaintiffs in the case of Kadrey et al. vs. Meta have filed a motion accusing the tech giant of knowingly utilizing copyrighted works in the development of its AI models, which raises significant questions about intellectual property rights in the age of artificial intelligence.

Background of the Case

The plaintiffs, including well-known author Richard Kadrey, recently submitted their “Reply in Support of Plaintiffs’ Motion for Leave to File Third Amended Consolidated Complaint” to the United States District Court for the Northern District of California.

The motion alleges that Meta systematically acquired copyrighted datasets through torrenting and stripped copyright management information (CMI) from these works, particularly from the infamous shadow library LibGen.

Incriminating Evidence Against Meta

Recent court documents suggest that Meta’s senior leaders were aware of incriminating practices. Notably, it is alleged that Meta CEO Mark Zuckerberg explicitly approved the use of the LibGen dataset, despite internal objections from the company’s AI executives.

An internal memo from December 2024 indicated widespread awareness within Meta of the pirated status of the LibGen dataset, igniting debates over its ethical and legal implications. Further documentation reveals hesitance from top engineers regarding the use of corporate resources for potentially illicit activities.

Stripping Copyright Management Information

Internal communications have surfaced suggesting that after acquiring the LibGen dataset, Meta intentionally removed copyright identifiers from the associated works. This practice is central to the plaintiffs’ claims of copyright infringement.

According to Michael Clark’s deposition, a corporate representative for Meta, the company employed scripts designed to erase any identifying information about the works’ copyrighted status, including keywords like “copyright” and “acknowledgements.” This action was reportedly aimed at preparing the dataset for training its Llama AI models.

An Ethical Dilemma

The allegations cast Meta in the light of a company knowingly engaging in a piracy scheme through torrenting. Emails exchanged between Meta engineers highlight concerns regarding the optics of downloading pirated datasets from corporate devices, with one engineer noting that “torrenting from a [Meta-owned] corporate laptop doesn’t feel right.”

Delayed Discovery and Bad Faith Claims

Legal counsel for the plaintiffs argues that as recently as January 2024, Meta had actively torrented data from LibGen. Documents indicate that hundreds of pertinent records were initially withheld during early discovery, which the plaintiffs view as bad-faith efforts to obstruct access to critical evidence.

Zuckerberg’s Deposition Insights

During a December 17, 2024 deposition, Zuckerberg reportedly acknowledged the questionable nature of Meta’s activities, admitting they could raise “lots of red flags” and expressing that such practices “seem like a bad thing.” However, responses regarding the broader scope of Meta’s AI training remained vague.

Expanding Legal Claims

The case, initially framed as an intellectual property infringement action by authors and publishers, is evolving with the plaintiffs seeking to introduce two significant additional claims: a violation of the Digital Millennium Copyright Act (DMCA) and a breach of the California Comprehensive Data Access and Fraud Act (CDAFA).

The DMCA allegations assert that Meta knowingly removed copyright protections to obscure unauthorized uses of copyrighted texts in its Llama models, with the plaintiffs claiming that such actions hindered copyright holders’ ability to identify infringements.

CDAFA Allegations and the Nature of Torrenting

The CDAFA allegations pertain to Meta’s methods of acquiring the LibGen dataset, suggesting the company engaged in torrenting to obtain copyrighted data without authorization. Internal discussions among Meta engineers raised concerns over the legality of their actions, highlighting an awareness of the potential legal ramifications.

Broader Implications for AI and Copyright Law

At the crux of this ongoing legal battle lies the growing tension between copyright law and AI technology. Plaintiffs contend that Meta’s practices deny rightful compensation to copyright owners, allowing the company to build AI systems like Llama on the creative efforts of authors and publishers.

Impact on AI Legislation

The allegations come at a time of heightened scrutiny on generative AI technologies, with various companies, including OpenAI and Google, facing similar concerns regarding their use of copyrighted content for model training. Courts in both the US and UK are navigating the complexities of AI’s implications for rights management.

Future of Copyright in the Age of AI

The unfolding case of Kadrey et al. vs. Meta could ultimately reshape legal precedents regarding AI development in the US and beyond. As copyright law struggles to keep pace with rapid technological advancements, there is an urgent need for clearer guidelines to protect the rights of creators and innovators alike.

For Meta, these allegations pose a reputational risk, especially as the company aims to solidify its position in the AI landscape. The reliance on pirated data may jeopardize its aspirations in a field where ethical considerations are paramount.

The outcome of this case could have far-reaching consequences for the future intersection of AI development and copyright law, as both creators and tech companies await clarity in an evolving regulatory environment.

Conclusion

As the legal landscape surrounding AI continues to evolve, the implications of the Kadrey et al. vs. Meta case may serve as a pivotal moment in determining the boundaries of copyright law and its application to emerging technologies.

Questions and Answers

1. What is the main allegation against Meta in this case?

Meta is accused of knowingly using copyrighted works in the development of its AI models, specifically through the acquisition of datasets from the pirated library LibGen.

2. Who are the plaintiffs in this lawsuit?

The plaintiffs include author Richard Kadrey and other authors and publishers who claim violations of their intellectual property rights.

3. What specific practices are being contested?

The plaintiffs allege that Meta stripped copyright management information from the works and engaged in torrenting to obtain the datasets without permission.

4. What potential legal claims are being introduced in this case?

The plaintiffs are seeking to add claims of violation of the Digital Millennium Copyright Act (DMCA) and breach of the California Comprehensive Data Access and Fraud Act (CDAFA).

5. How might this case impact the future of AI and copyright law?

The outcome of the case could establish important legal precedents concerning the intersection of AI technology and copyright law, shaping how future AI models are developed in relation to intellectual property rights.

(Photo by Amy Syiek)

See also: UK wants to prove AI can modernise public services responsibly

Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is co-located with other leading events including Intelligent Automation Conference, BlockX,Digital Transformation Week, and Cyber Security & Cloud Expo.

Explore other upcoming enterprise technology events and webinars powered by TechForge here.