Meta is embroiled in a copyright lawsuit that has exposed internal communications regarding its AI model development, particularly the alleged use of pirated data for training its Llama models.

The lawsuit, brought forth by notable authors including Sarah Silverman and Ta-Nehisi Coates, claims that Meta cross-referenced pirated books with licensed works to explore potential licensing agreements.

Internal discussions revealed that Meta employees sought to filter out copyright indicators from LibGen materials, indicating a deliberate effort to obscure the origins of the data used.

Documents also show that Meta took steps to minimize legal liability by suggesting the removal of copyright headers and author lists from the LibGen dataset.

The outcome of the Kadrey v. Meta case could have significant implications for the future of AI development and intellectual property rights, as courts grapple with these complex issues.

Recent filings suggest that Meta downloaded pirated e-books from Z-Library for training purposes as recently as April 2024, despite Z-Library facing legal actions for copyright violations.

Meta's consideration of using Library Genesis (LibGen) as a source for training data has raised alarms about its legality, as the site is notorious for providing access to copyrighted materials.

This lawsuit could set a precedent for ongoing legal challenges faced by AI companies regarding copyright infringement and the ownership of digital content.

Amidst scrutiny over its practices, Meta is also cutting approximately 3,600 employees, representing 5% of its workforce, as part of a restructuring effort.

Meta's defense for using copyrighted material hinges on claims of legal fair use, a position that aligns with similar arguments made by other AI companies facing lawsuits.

The SPAR3D model, recently unveiled by Stability AI, can generate complete 3D objects from a single image in under one second, showcasing advancements in AI technology.

During a deposition, CEO Mark Zuckerberg claimed unfamiliarity with LibGen, despite internal concerns about its usage and the legal implications surrounding it.