xAI Sues Grok User Terry Harwood Over Alleged Child Abuse Deepfakes
Elon Musk’s artificial intelligence company xAI has taken the unusual step of suing one of its own Grok chatbot users, accusing South Carolina man Terry Wayne Harwood of using the tool to generate child sexual abuse material and non-consensual sexualised deepfakes of adults. The complaint, filed Tuesday in the US District Court for the Northern District of Texas, marks one of the first known instances of an AI company legally targeting a user for the exact kind of abusive content it has spent the past year fending off lawsuits over.
Read also: OpenAI’s First Hardware: A Screenless, Movable ChatGPT Speaker Companion
What xAI Alleges Harwood Did With Grok
xAI alleged in the lawsuit that Harwood violated its terms of service “to convert non-sexual photographs into sexually explicit images without the photograph subjects’ knowledge or consent.” The suit claims Harwood created multiple xAI accounts and entered “misleading prompts” to generate explicit deepfakes of both minors and adults.
The lawsuit further alleges that Harwood designed prompts it characterized as “misleading” in order to “circumvent” what it described as “built-in safeguards” meant to prevent the creation of explicit material depicting minors. “Defendant breached the xAI Terms of Service and Acceptable Use Policy by leveraging Grok to generate non-consensual sexually explicit images and CSAM,” the suit adds.
Harwood, along with three other men, was arrested earlier this year in South Carolina on multiple counts of alleged sexual exploitation of a minor. He was arrested in February, and contact information for him was not immediately available, according to Reuters.
xAI Wants Damages and a Permanent Grok Ban
xAI asked the court for an unspecified amount of monetary damages and a court order permanently blocking Harwood from using Grok. The company framed the case in stark terms, arguing in its filing that “Defendant’s actions were a calculated scheme to weaponize Plaintiff’s tool for criminal ends, exposing real victims to profound and lasting harm, while exposing Plaintiff to significant legal risk and reputational damage.”
Beyond the individual case, xAI used the filing to publicise its broader enforcement record. In the lawsuit, xAI claims that it “has suspended 52,222 accounts and made 73,604 reports to (National Center for Missing & Exploited Children) in 2026, resulting in (at least) 244 arrests.”
From Defendant to Plaintiff: xAI’s Legal Reversal
The case is one of the first brought by an AI company against one of its own users for allegedly using the system to generate explicit material. That framing is notable given how much of the past seven months has seen xAI itself sitting in the defendant’s chair rather than the plaintiff’s.
Earlier this year, Grok came under fire for a tool that allowed “digital undressing,” including AI-generated images of children, and after users began noticing a surge in non-consensual, sexualised deepfakes late last year, Musk and xAI said they would take action against such illegal use of the AI. Musk himself wrote in a January 6 social media post quoted in the suit that “Anyone using Grok to make illegal content will suffer the same consequences as if they upload illegal content.”
That controversy has since spawned a wave of separate litigation against the company. A class action originally filed in March on behalf of teenage victims in Tennessee has since been amended to add two more plaintiffs and a second AI defendant, Stability AI, with one new claimant alleging her stepfather generated thousands of abusive images from a single childhood photo of her. Influencer Ashley St. Clair also sued xAI in New York, alleging Grok was used to create sexualised images of her, including from photos taken when she was a minor, while the city of Baltimore has separately sued the company over alleged consumer-protection violations.
Australia’s eSafety Commissioner Has Been Watching Grok Since January
The Harwood case lands against a backdrop of regulatory pressure that reached Australian shores well before this lawsuit. Australia’s eSafety Commissioner, Julie Inman Grant, flagged concerns after an “edit image” feature added to Grok late last year was found to let users strip clothing from photos of real people, including children, without their consent.
Inman Grant said “Since late 2025, eSafety has received a doubling of reports relating to the use of Grok to generate sexualised images without consent,” adding that “Some reports relate to images of adults, which are assessed under our image-based abuse scheme, while others relate to potential child sexual exploitation material.” She noted that “Australia’s enforceable industry codes and standards require online services to implement systems and processes to safeguard Australians from illegal and restricted material, including child sexual exploitation material, whether it’s AI-generated or not.”
Australia has already shown it is willing to punish individual offenders in this space. eSafety previously secured the country’s first deepfake penalty, a federal court fine against a Gold Coast man for distributing deepfake pornography of prominent Australian women, a precedent that may be watched closely as US authorities and xAI itself now test individual liability for Grok misuse through the courts.
A Test Case for AI Platform Accountability
Legal experts have increasingly framed AI-generated sexual abuse imagery not just as a privacy violation but as a product-safety issue, since harm can occur the instant an image is generated regardless of how widely it is later shared. Suing an individual user directly gives xAI a chance to argue in court that responsibility for such abuse sits squarely with the people who deliberately bypass safeguards, rather than with the design of Grok itself, a distinction that will likely feature heavily in the company’s defence against the separate class actions still working their way through federal court in California.
For now, the Harwood suit remains in its early stages, and it is unclear whether he has yet responded in court. But its filing signals that xAI intends to pair its public safety pledges with active litigation against users who test the platform’s limits, even as it continues to fight allegations that Grok’s own guardrails were inadequate in the first place.
Join the Conversation