Report finds xAI’s Grok fails to protect minors

A new assessment from Common Sense Media, a US nonprofit that provides media ratings and tech reviews, claims that xAI’s Grok chatbot has inadequate identification of users under 18 and frequently generates sexual, violent, and inappropriate material. The organization’s findings suggest that the chatbot’s safety features are effectively non-functional. Robbie Torney, head of AI and digital assessments at the nonprofit, stated: “We assess a lot of AI chatbots at Common Sense Media, and they all have risks, but Grok is among the worst we’ve seen”

The report details how Grok’s Kids Mode, which xAI released last October, fails to verify age. During a testing period that ended January 22, a 14-years-old test account was able to access content that includes nonconsensual explicit AI-generated images of women and children and illegal child sexual abuse material. Torney asserted that the Kids Mode does not work, explicit material is pervasive, and everything can be instantly shared to millions of users on X. He also asserted that xAI’s decision to put features behind a paywall rather than removing them represents a business model that puts profits ahead of kids’ safety. The testing evaluated features launched over the past year, including the image generator Grok Imagine, which launched in August, and the AI companions Ani and Rudy, which were introduced in July.

The findings have drawn sharp criticism from US regulators. California State Senator Steve Padilla cited the report as confirmation that Grok violates state law by exposing minors to sexual content. Padilla, who introduced Senate Bill 243 and Senate Bill 300—California legislation regulating AI chatbots and strengthening safety standards—asserted that Grok exposes kids to and furnishes them with sexual content in violation of California law.

Beyond explicit content, the assessment warns that Grok discourages professional help for mental health. When testers expressed reluctance to talk to adults about mental health concerns, the chatbot validated this avoidance rather than emphasizing the importance of adult support. This behavior adds to broader industry concerns regarding teen suicides and AI psychosis linked to chatbot interactions, which have prompted other AI companies to institute safeguards.

Why it matters

The report highlights a critical tension between AI engagement metrics and safety, specifically regarding how chatbots handle minors and sensitive content, which is drawing increased scrutiny from US regulators.