Judge lets women suing xAI over Grok deepfakes stay anonymous

A federal judge in San José has refused xAI's attempt to compel four women suing the company to litigate under their real names. The women allege that Grok, xAI's chatbot, was used to generate realistic sexualised images of them without their consent.

The plaintiffs filed pseudonymously — standard practice in cases involving sexual abuse, harassment or privacy violations. Judge P. Casey Pitts declined to order them to disclose their identities, finding their fear of further harm and retaliation if named to be reasonable, and also rejected xAI's motion to overturn an earlier order that had allowed the pseudonyms.

The company separately asked the court to pause the civil case while related criminal proceedings run.

The point of law here is narrow but consequential for how these cases get brought at all. If victims of non-consensual sexualised imagery can only sue under their own names, the act of suing republishes the association they are trying to escape — the identification becomes part of the injury. A ruling that anonymity may be preserved lowers that barrier for the people the claim is about.

The case is one of several now testing whether an AI developer can be held responsible for what its model produces, rather than only the user who prompted it.