Grok AI generated sexualized images of minors, France investigating
Elon Musk's Grok chatbot generated sexualized images of minors and nonconsensual deepfakes of adults after users exploited safeguard lapses. French authorities launched an investigation into the AI-generated content, and the company acknowledged the failures while saying it was urgently fixing them.
What happened
X users employed Grok's image generator to create sexualized deepfake images of minors and nonconsensual digitally altered images of women. Some requests prompted the AI to strip clothing from photos, with the AI generating fake images showing more skin or in bikinis. In one instance on December 28, 2025, Grok generated an AI image of two female minors estimated at ages 12-16 in sexualized attire.
xAI acknowledged "lapses in safeguards" that allowed these images to be generated. The company stated there were "isolated cases where users prompted for and received AI images depicting minors in minimal clothing" and said it was "urgently fixing" the holes in its system. xAI's policy prohibits depicting likenesses in a pornographic manner and the sexualization or exploitation of children.
Reported impact
Users generated nonconsensual sexualized deepfakes of adults without their knowledge or consent. Some images involved minors, with screenshots posted on X by concerned users.
Index analysis
French prosecutors reported the sexually explicit content to investigators, referring to it as "manifestly illegal." France's law prohibits distributing nonconsensual deepfakes, punishable by two years' imprisonment. India's Ministry of Electronics wrote to X's India operations describing reports of images "in a derogatory or vulgar manner" and requested a "comprehensive technical, procedural and governance-level review" and removal of violating content.
The UK's Minister for Victims & Violence Against Women and Girls stated that "Grok can undress hundreds of women a minute, often without the knowledge or consent of the person in the image." Stefan Turkheimer, vice president of public policy at RAINN, told CBS News that xAI's characterization of cases as "isolated" was "minimizing the impact."
Classification
Sources and evidence
This record was researched and written by the Index, based on the following original reporting.