A group of girls has filed a class action lawsuit against Elon Musk’s xAI, alleging that its chatbot Grok was used to create sexually explicit images of minors without effective oversight, Ars Technica and Futurism report.
In the most severely described case, the lawsuit claims a man provided Grok with a photograph of his stepdaughter taken when she was 11 years old, generating around 7,000 explicit images of the girl. The chatbot’s safety systems flagged almost nothing as problematic: a report to the National Center for Missing and Exploited Children (NCMEC) was sent only after one specific prompt, after which authorities were notified.
The lawsuit alleges that xAI then "obstructed" the investigation "at every turn", refusing to hand over key user identification data, including the IP address. Police did locate the man and discovered a large collection of generated images. Two days after being released on bail, he shot himself. The girl is referred to in the lawsuit as Jane Doe 4. "They had everything needed to help police stop the responsible person and secure justice. Instead, they remained silent," she said, Ars Technica reports.
Plaintiffs also cite NCMEC data stating that 90 per cent of reports received from xAI were unactionable because they contained insufficient information about the reported users.
Grok’s ability to create sexualised deepfakes caused a stir as early as December and January, when it generated tens of thousands of fake nude photographs of real women and children. Company Copyleaks then estimated that one such image was created every minute, and the Centre for Countering Digital Hate later calculated that around three million had been produced, of which more than 23,000 depicted children.
Musk denied that Grok was used to create child sexual abuse material. Following criticism, the company first restricted image generation to subscribers and then announced stricter safeguards against creating sexual images of real people.
