The federal judge has dismissed some of the lawsuits of AI -related lawsuits against the meta, but allows them to move forward.
Kadrey VS. The authors, including Richard Kadrey, Sarah Silverman, and Ta-Nehisi Coates in Meta, insisted that Meta used books to train the LLAMA AI model to infringe intellectual property rights and the company removed copyright information from the book to hide the charges.
Meta, meanwhile, insisted that training was qualified for fair use, and the authors insisted that they should be dismissed because they were sued. In last month’s court, Vince Chhabria’s US judge seemed to indicate that he was opposed to dismissal, but he also criticized the author’s “best” investigation.
In Friday’s ruling, Chabria alleged that copyright infringement was “a specific injury to the standing,” and the authors also argued that the meta was intentionally eliminating CMI (copyright management) and hiding copyright infringement.
“This argument is a reasonable, especially strong reasoning that META has been trained for copyrighted data by removing CMI to prevent LLAMA from output of CMI,” said CHABRIA.
The judge, however, dismissed the author’s claims related to California comprehensive computer data access and fraud (CDAFA). Because he did not claim that META has access to a computer or server.
The lawsuit has already provided some glimpse of the Meta’s approach to copyrights, and the plaintiff’s court submissions that Mark Zuckerberg has granted the authority to train the model using copyrighted works on the LLAMA team, and that other META team members discussed the use of legally suspicious contents in AI education.
The court is currently suspending many AI copyright lawsuits, including New York Times litigation against Openai.
