Meta Wins Lawsuit Over AI Training Data | Copyright & AI

Meta Wins Copyright Lawsuit Over AI Training

A federal judge has ruled in favor of Meta on Wednesday, dismissing a lawsuit filed by 13 authors, including Sarah Silverman. The suit claimed that Meta illegally utilized their copyrighted works to train its AI models.

Fair Use Doctrine Applied

Judge Vince Chhabria granted a summary judgment, resolving the case without a jury trial. He determined that Meta’s training of its AI models using copyrighted books qualified as “fair use” under copyright law, and was therefore permissible.

This decision follows a similar ruling just days prior, where a federal judge also sided with Anthropic in a comparable legal challenge. These outcomes are increasingly favorable for the technology sector.

Limited Scope of the Ruling

Despite these victories, the judges emphasized that the rulings are not broad endorsements of all AI training practices. Both cases were specifically limited in their scope.

Judge Chhabria clarified that his decision doesn't establish the legality of all AI model training involving copyrighted material. He stated the plaintiffs in this instance “presented inadequate arguments” and lacked sufficient evidence to support stronger claims.

“This ruling doesn’t establish that Meta’s use of copyrighted materials for AI language model training is lawful,” Judge Chhabria explained. He further added, “In cases resembling Meta’s, plaintiffs may often succeed, particularly when presenting more thoroughly developed evidence regarding the market impact of the defendant’s actions.”

Transformative Use and Market Impact

The judge found Meta’s use of copyrighted works to be transformative. This means the AI models did not simply replicate the authors’ books.

Crucially, the plaintiffs were unable to demonstrate that Meta’s use of the books negatively impacted the market for those authors’ work – a key consideration in copyright infringement cases.

“The plaintiffs did not offer any substantial evidence of market dilution,” Judge Chhabria noted.

Ongoing Legal Battles

While both Anthropic and Meta’s cases involved training AI models on books, numerous other lawsuits are currently active. These cases target technology companies for training AI models on diverse copyrighted content.

• The New York Times is pursuing legal action against OpenAI and Microsoft for utilizing news articles.
• Disney and Universal are suing Midjourney for training AI models on films and television programs.

Industry-Specific Considerations

Judge Chhabria highlighted that fair use defenses are highly dependent on the specifics of each case. He suggested that certain industries may have more compelling fair use arguments than others.

“Markets for specific types of works, such as news articles, may be particularly susceptible to indirect competition from AI-generated content,” Chhabria observed.