May 8 (Reuters) – A California federal judge said he was inclined to green-light a copyright lawsuit against Stability AI, Midjourney and other companies accused of misusing visual artists’ work to train their artificial intelligence-based image generation systems.
Orrick issued the “tentative rulings” on Tuesday in advance of a court hearing scheduled for Wednesday.
Representatives for the companies did not immediately respond to requests for comment on Wednesday. Matthew Butterick, an attorney for the artists, declined to comment.
Orrick said that the claims likely had enough merit to continue and should be tested at the next stage of the case against “various direct and induced infringement theories and precedent under the Copyright Act.”
AI companies have argued that they make fair use of copyrighted data to train their systems. Orrick’s opinion did not address fair use, which is likely to be a crucial question for AI copyright cases in the future.
Orrick also said that he was likely to dismiss some of the artists’ related claims but allow their allegations that the companies violated their trademark rights and falsely implied that they endorsed the systems.
The case is Andersen v. Stability AI, U.S. District Court for the Northern District of California, No. 3:23-cv-00201.
For the artists: Joseph Saveri of Joseph Saveri Law Firm; Matthew Butterick; Brian Clark of Lockridge Grindal Nauen
For Stability: Joseph Gratz of Morrison & Foerster
For Midjourney: Angela Dunning of Cleary Gottlieb Steen & Hamilton
For DeviantArt: Andy Gass of Latham & Watkins
For Runway: Paven Malhotra of Keker Van Nest & Peters
Read more:
Sign up here.
Reporting by Blake Brittain in Washington
Our Standards: The Thomson Reuters Trust Principles.
