A federal judge has decided to allow a copyright lawsuit involving artificial intelligence to proceed against Meta, although he dismissed certain aspects of the case. In the lawsuit titled Kadrey vs. Meta, notable authors such as Richard Kadrey, Sarah Silverman, and Ta-Nehisi Coates have accused Meta of infringing on their intellectual property rights by using their books to train its Llama AI models. They also allege that Meta removed copyright information from their works to obscure this infringement.
In response, Meta asserts that its use of the authors’ works qualifies as fair use and argues that the authors do not have the legal standing to pursue the case. During a court hearing last month, U.S. District Judge Vince Chhabria expressed skepticism about dismissing the case outright, although he did critique what he considered excessive rhetoric from the authors’ legal representatives. In his ruling on Friday, Judge Chhabria emphasized that the authors’ claims of copyright infringement represent a tangible injury that provides them with standing to sue. He noted that they had sufficiently claimed that Meta intentionally removed copyright management information (CMI) to hide its copyright violations. Chhabria stated, “Taken together, these allegations raise a ‘reasonable, if not particularly strong inference’ that Meta removed CMI to try to prevent Llama from outputting CMI and thus revealing it was trained on copyrighted material.
” However, the judge dismissed the authors’ claims under the California Comprehensive Computer Data Access and Fraud Act (CDAFA), stating that they failed to demonstrate that Meta had accessed their computers or servers, as they only accused the company of accessing their data in the form of their books. This lawsuit has begun to shed light on Meta’s stance on copyright issues, with court documents revealing that CEO Mark Zuckerberg allegedly authorized the Llama team to use copyrighted materials for training and that other Meta employees discussed the use of potentially legally dubious content for AI training.
The legal landscape around AI and copyright is rapidly evolving, with several lawsuits, including The New York Times’ case against OpenAI, currently under consideration by the courts. These cases are expected to have significant implications for how AI companies handle copyrighted material and the boundaries of fair use in the context of AI training.
