Earlier this month, St. Clair filed a lawsuit against Musk’s AI company in response to a flood of deepfakes made by xAI’s chatbot, Grok, and shared on X, the social media platform Musk owns. She told Fortune earlier in January that X users were turning pictures on her X profile into sexualized AI-generated images, including some she said depicted her as a minor. 

“There were pictures of me with nothing covering me except a piece of floss with my toddler’s backpack in the background, and photos of me where it looks like I’m not wearing a top at all,” she said. “I felt so disgusted and violated.”

The dispute between St. Clair and X is just one part of a wider wave of outrage and backlash after xAI’s Grok AI model was used for a widespread, nonconsensual “digital undressing” campaign that began in late December. Many users discovered they could tag Grok on the X platform and ask it to edit images from users’ profiles. According to research from the Center for Countering Digital Hate, over a span of 11 days—from Dec. 29, 2025, through Jan. 8, 2026—Grok generated approximately 3 million sexualized images.

In the suit, St. Clair, who is Jewish, also claims the AI chatbot generated an image that put her “in a string bikini covered with swastikas.” The more St. Clair spoke out about the issue, she said, the more she was targeted with the images.

The legal fight got unpleasant quickly. After being notified of the action, Musk’s company almost immediately countersued the conservative influencer, accusing her of violating the platform’s terms of service, which state any disputes with X must be brought in Texas. Representatives for Musk did not respond to specific questions from Fortune.

Musk has typically preferred to fight his legal battles in Texas. In 2024, X updated its terms of service to require that all federal lawsuits against the company be filed in the Northern District of Texas, despite its headquarters being in Bastrop, which falls under the Western District of Texas.

Musk and the company have not stated their reasons for that stipulation. But 10 of the Northern District’s 11 active judges were appointed by Republican presidents, and unlike most federal courts, the district assigns cases based on which division they’re filed in rather than through random assignment. Ethics experts have also raised potential conflict‑of‑interest concerns because Judge Reed O’Connor, who has overseen several lawsuits involving Musk’s platform X, has owned Tesla stock. O’Connor has sometimes declined to recuse himself, though he at least once stepped aside after his holdings drew scrutiny.

“It’s really just ridiculous,” St. Clair said of the countersuit, arguing that X’s terms of service don’t apply since she’s suing over images created of her without her consent, not over her own use of the platform.

St. Clair says she’s now trying to fight the countersuit, block any venue change to Texas, and fight to establish greater responsibility from generative AI platforms.

“It’s about holding these platforms accountable and holding these very new generative AI tools accountable when they’re launched without any regard for the damage they’ve done, damage that was completely foreseeable,” she said. “I will never get the images of not only myself, but of other people’s children and other women that I had to see, out of my head.”

XAI released an update of the chatbot earlier this month and claimed that Grok’s X account was no longer allowed to edit images of real people in revealing clothing. St. Clair says that users have found workarounds, including using the stand-alone Grok app, and that the images are still being distributed on the X platform.

“It’s still happening. People are still able to produce these images on the stand-alone Grok app and website,” she said. “There are also images of me being burned that are being produced. Still, provocative images of me are being produced. It hasn’t stopped.”

European nonprofit AI Forensics also found that Grok was still generating sexualized images of individuals despite X’s restrictions. The researchers found users bypassing the ban by accessing Grok directly through its website rather than through X, or by using Grok Imagine, the AI’s video and image generation tool.

St. Clair has faced blowback from her fight with xAI and Musk: Comments on X regularly accuse her of being a vindictive ex-girlfriend who’s after Musk’s money; she was stripped of her “verified” status on X and the ad-revenue-sharing scheme that came with it; and in one now-deleted X post, Musk even threatened to sue for full custody of their child after St. Clair made comments expressing regret for her previous stance on trans rights.

Grok has sparked global regulatory scrutiny

X’s “nudifying” trend that targeted St. Clair has sparked global concern in part owing to the proliferation of sexualized images of children. According to research from the Center for Countering Digital Hate, of the 3 million sexualized images produced by Grok in an 11-day period, 23,000 appeared to depict children, reflecting an estimated average pace of one new sexualized image of a minor every 41 seconds. These included a selfie uploaded by a schoolgirl who was “undressed” by Grok, and an image of six young girls wearing micro bikinis, generated by Grok. 

“It’s really hard not to conclude that this is a disaster of [Elon Musk’s] own making,” Imran Ahmed, founder and CEO of the Center for Countering Digital Hate, told Fortune. “Musk was very clear that in his own personal postings, he mocked people who were concerned about how it was being abused, and seemed to make light of it, rather than take it seriously and seek to remedy the situation for almost two weeks.”

Grok is also facing a class action lawsuit in California over the sexualized deepfakes, as well as a formal probe by the EU Commission. If X is found to have breached the EU’s Digital Services Act, the company could face fines of up to 6% of its global annual revenue. 

“The EU actions are fantastic. I think it’s really a step in the right direction, and I hope more people and leaders follow,” St. Clair said. The Trump administration has taken multiple actions against what it views as European “censorship” and has a particular issue with the Digital Services Act. In December, in one of the administration’s more dramatic moves, it barred five European nationals, including Ahmed, from entering the U.S., for allegedly pressuring tech firms to censor American viewpoints. 

On Monday, the EU regulator said it would assess whether “manipulated sexually explicit images” generated by Grok have been shown to users in the bloc, and warned it may “impose interim measures” if X refuses to implement meaningful adjustments. Similar investigations are underway in Australia, France, and Germany, while X was temporarily banned in Indonesia and Malaysia.

“AI platforms don’t have the same protections from litigation that social media companies do, and this is an important test of whether or not negligence law and prompt design law can be applied to AI platforms,” Ahmed said. “If it’s found that they are liable, it will likely change the AI chatbot industry forever.”