Meta Accused of Torrenting 81.7TB of Pirated Books to Train AI Models

Unsealed emails reveal Meta employees knowingly used shadow libraries like LibGen and Z-Library for AI training, raising serious legal and ethical concerns.

• Authors, including Sarah Silverman and Richard Kadrey, have filed a lawsuit accusing Meta of using pirated books to train its AI models, including Llama.
• Court documents reveal Meta downloaded at least 81.7 terabytes of data from shadow libraries such as LibGen and Z-Library, with internal emails acknowledging the use of pirated content.
• Meta employees expressed ethical and legal concerns about torrenting copyrighted materials, with some suggesting measures to conceal the activity, such as using VPNs and operating in 'stealth mode.'
• Evidence suggests Meta executives, including Mark Zuckerberg, were aware of the use of pirated materials, and the company attempted to mask its torrenting activities by using servers outside its main network.
• The plaintiffs are seeking access to Meta's torrenting logs, peer-sharing records, and datasets for future AI models, arguing that the magnitude of the alleged piracy warrants further investigation.

3 Articles

TechSpot TweakTown PC Gamer