SAN FRANCISCO – A federal judge has dismissed a lawsuit brought by 13 authors, including Sarah Silverman and Junot Díaz, who claimed that Meta’s AI model training violated their copyrights.
Breaking: Judge Backs Meta in Copyright Dispute
On Wednesday, U.S. District Judge Vincent Chhabria ruled that Meta’s use of a dataset comprising nearly 200,000 books, including those of the plaintiffs, to train its Llama language model constitutes “fair use.” This decision aligns with a similar ruling earlier in the week regarding Anthropic’s language model, Claude.
Immediate Impact
The court’s decision is seen as a significant victory for Meta and the broader AI community, as it underscores the legal framework allowing the use of copyrighted material in AI training under fair use provisions. A Meta spokesperson expressed satisfaction with the ruling, highlighting the role of open-source AI models in driving innovation and creativity.
“Open-source AI models are powering transformative innovations, productivity, and creativity for individuals and companies, and fair use of copyright material is a vital legal framework for building this transformative technology,” a Meta spokesperson stated.
Key Details Emerge
Judge Chhabria rejected the authors’ assertion that Meta’s actions amounted to “unmitigated piracy.” He noted that Llama’s inability to replicate more than 50 consecutive words makes it a “transformative” use of the original works. However, the judge acknowledged the potential threat AI poses to the market for original works if it leads to the creation of numerous low-cost imitations.
“What copyright law cares about, above all else, is preserving the incentive for human beings to create artistic and scientific works,” Judge Chhabria wrote.
Industry Response
The ruling has garnered varied reactions from the legal and creative communities. Boies Schiller Flexner LLP, representing the plaintiffs, expressed disagreement with the court’s decision. The firm highlighted the court’s acknowledgment that using copyrighted works without permission generally violates the law, yet noted the ruling favored Meta.
“Despite the undisputed record of Meta’s historically unprecedented pirating of copyrighted works, the court ruled in Meta’s favor. We respectfully disagree with that conclusion,” a spokesperson for the firm stated.
Background Context
The lawsuit, filed by a group of authors including Richard Kadrey, Christopher Golden, and Ta-Nehisi Coates, alleged that Meta utilized “shadow libraries” to acquire millions of pirated books. The authors claimed that Meta’s engineers used BitTorrent to download and sometimes reupload the data, further infringing on their copyrights.
What Comes Next
While the judge dismissed the AI training claim, the issue of Meta’s alleged use of torrenting remains unresolved. The plaintiffs’ legal team has not yet indicated whether they will appeal the decision, leaving open the possibility of future legal challenges.
“Meta introduced evidence that its copying hasn’t caused market harm,” Judge Chhabria wrote. “The plaintiffs presented no empirical evidence to the contrary… All the plaintiffs presented is speculation.”
Future Implications
This ruling could set a precedent for future cases involving AI and copyright laws, though Judge Chhabria emphasized that his decision is specific to the facts presented in this case. As the legal landscape around AI and intellectual property continues to evolve, stakeholders across industries are closely monitoring developments.
The authors involved in the lawsuit, including Andrew Sean Greer, David Henry Hwang, and Jacqueline Woodson, have yet to comment on their next steps. The decision may influence how companies approach the use of copyrighted materials in AI development moving forward.
