Key Takeaways

• xAI is moving to strip four Grok deepfake victims of their courtroom anonymity, forcing them to either go public or drop their case entirely
• British MP Jess Asato has filed a separate lawsuit against xAI at the High Court, using her public profile to fight for accountability on behalf of women who cannot
• Legal experts warn that forcing real names in privacy cases deters victims from pursuing litigation at all, which may be exactly the intention
• This is not a moderation failure. Grok's ability to generate non-consensual sexualised images was a design choice, and the cover-up is now part of the pattern

xAI Wants Grok's Deepfake Victims to Choose: Go Public or Lose Your Case

When we covered the EU's condemnation of Grok's non-consensual image generation back in January, we wrote that how platforms are built and handled matters just as much as the technology itself. Five months later, that argument has become harder to ignore. Four people suing Elon Musk's AI firm under pseudonyms are now facing a grim choice: reveal their real names, or drop the lawsuit. This isn’t a procedural footnote. It is a strategy.

What xAI Is Doing in Court

xAI's position is that there’s insufficient evidence of specific future harm to justify continued anonymity, particularly since the images in question remain under seal. That framing sounds almost reasonable until you read what the plaintiffs actually submitted. 

They filed affidavits in late May describing severe emotional distress and fears of doxxing, online harassment, and the creation of additional deepfakes if their identities became public. One plaintiff disclosed that law enforcement had informed her about alleged child sexual abuse material created from images of her as a child.

The plaintiffs' legal team has characterized xAI's motion as a deliberate intimidation tactic. It’s difficult to argue otherwise. Legal expert Danielle Citron has highlighted that requiring real names in privacy cases may deter victims from pursuing litigation at all. That deterrent effect isn’t a side consequence of xAI's motion. It may well be the point.

The Damage Already Done in Public

British MP Jess Asato, a member of Prime Minister Keir Starmer's Labour Party, announced on June 3rd that she is suing xAI after the Grok platform was used to create fake sexualised images of her. Unlike the four anonymous plaintiffs in the US case, Asato isn’t hiding. She is using her public profile as a weapon, and she is doing it on behalf of everyone who can’t.

After she condemned Grok in January, users created and shared fake images of her in a bikini, and a video showing her being chloroformed and prepared for a sexual assault. That is what happened to a sitting lawmaker who spoke publicly about the problem. It should be obvious why others would want to remain anonymous.

Asato filed her claim at the High Court in England for breaches of data protection law and misuse of private information, seeking damages, a formal acknowledgment that what happened was illegal, and an order requiring xAI to stop all further illegality. Her legal team put it plainly: "This is one of the first claims to test liability for the design of an AI system, and we hope it will make it clear to AI developers that safety cannot be an afterthought."

This Is a Design Problem, Not a Misuse Problem

xAI's defenders will argue that abusers are responsible for abuse, not the platform. That’s technically true and practically useless. Asato herself said: "Grok created deepfake pornography and sexualised content which harmed thousands of women and children. Its ability is not an accident, nor misuse, it is a design choice by its creators."

In earlier proceedings, xAI claimed it was impossible to guarantee the platform couldn’t be used for abuse and that responsibility lay with malicious users. Meanwhile, in early February, even after xAI announced new curbs in mid-January, Grok continued to generate sexualised images of people, even when users explicitly warned that the subjects did not consent. The restrictions were cosmetic. The harm continued.

I don’t believe banning tools or restricting access by region solves anything. It just relocates the problem. What I do believe is that platforms have a responsibility to build safety into the product from the start, and to face meaningful consequences when they refuse to.

Privacy Isn’t a Technicality

There is something worth saying plainly about what it means to ask a deepfake victim to attach her name to a lawsuit. This is a woman who had pornographic images created of her without her consent. She didn’t choose to be part of this story. She is already dealing with the psychological wreckage of having her image weaponized. Now she’s being told that accessing the legal system requires her to accept permanent, searchable, public identification as a victim of sexual exploitation.

Courts generally allow pseudonymous filings when plaintiffs face legitimate risks of harm, harassment, or invasion of privacy. These are precisely the concerns deepfake victims cite. The fact that xAI is challenging this in cases involving non-consensual sexual imagery tells you everything about how seriously the company takes the harm it enabled.

Everyone has a right to exist online without their image being turned into a tool of harassment. Everyone has a right to pursue justice without surrendering that same privacy. A free internet means everyone gets to participate, not just the people powerful enough to absorb the consequences of speaking up.

The women and children targeted by Grok didn’t give up their right to safety by posting a photo. xAI built the machine that hurt them. The least it can do is let them fight back without forcing them to take off their masks first.