Arkansas Mother Sues xAI Alleging Grok Was Used to Turn Preschool Photos Into Abuse Material

A Bentonville mother says she got the news from a federal agent. According to a lawsuit filed Wednesday in Benton County Circuit Court, an officer with the U.S. Department of Homeland Security told her that photos taken of her daughter at preschool had been altered to create child sexual abuse material. The complaint alleges the tool used to do it was Grok, the AI system built by Elon Musk's xAI.

The suit, filed on behalf of the mother and her 5 year old daughter, who are identified only by pseudonyms, names X.AI Corp., X.AI LC and Bentonville photographer Russell Edward Bloodworth as defendants and seeks unspecified damages, the Northwest Arkansas Democrat-Gazette reported. According to the complaint, the school hired Bloodworth to take student photographs, and he photographed the girl four times while she attended between the ages of 2 and 4.

The allegations against the software are blunt. The lawsuit claims Bloodworth was able to upload ordinary photographs of minors to Grok, enter a few prompts and watch it generate thousands of new abuse images featuring the children's likenesses. It accuses xAI of negligently failing to build the safeguards, filters and other security mechanisms that would have stopped Grok from producing that material.

"No person should be allowed to do that to anyone else and no company should have a platform that allows it." Shane Wilkinson, attorney for the mother and daughter

None of this has been proven in court. Bloodworth, 43, faces 100 counts of pandering or possessing material depicting sexually explicit conduct involving a child and 100 counts tied to the unlawful creation or distribution of deepfake material, according to the Democrat-Gazette. He has pleaded not guilty. Bentonville police received a tip about him in June from the National Center for Missing and Exploited Children, the complaint says, and it alleges investigators found about 1,700 AI-generated abuse images and videos on his phone plus thousands more across more than 60 storage devices in his home.

This is not the first such case. KATV reported in July that another Arkansas family had sued xAI over Bloodworth's alleged use of Grok, with attorney Derek Potts saying more families had contacted his firm after police notified them. Wilkinson told the Democrat-Gazette that it was important for "these local cases" to stay in Benton County "where the crimes occurred."

xAI points at the user

xAI did not return the Democrat-Gazette's requests for comment. Its legal position, though, is on the record. In August the company sued Bloodworth in federal court in Texas, saying it wants to enforce its contractual rights and hold him accountable for "repeated deliberate, and unconscionable violations of the xAI Terms of Service and Acceptable Use Policy." The company has also said it has "zero tolerance for any forms of child sexual exploitation, non-consensual nudity, and unwanted sexual content."

That framing, a bad actor breaking the rules, sits awkwardly beside the record of Grok's image tools. The Center for Countering Digital Hate has said Grok produced 3 million images it described as sexualized between Dec. 29 and Jan. 8, including 23,000 depicting children. Potts told KATV the company marketed itself as "rebellious" compared with rivals.

The fallout is not limited to Arkansas. On the same day the new suit surfaced, the Bureau of Investigative Journalism reported that South Wales Police had reopened the case of Jess Davies, one of the women targeted last Christmas when explicit Grok images were being posted on X at a rate of up to 6,000 per hour, according to the Bureau. Police had closed her case in May. Davies, who the Bureau says is starting civil action against Musk's company, welcomed the decision but said it was "rather frustrating that victims have to fight so hard to be taken seriously and have their cases reviewed in the first place." The Bureau said SpaceXAI, formerly xAI, did not respond to its request for comment.

Courts will now decide how much of the responsibility belongs to the man accused of typing the prompts and how much, if any, belongs to the company that built the machine. For the families involved, the lawsuit alleges, the damage is already done.