Meta’s Celebrity-Voiced AI Chatbots Under Fire for Inappropriate Sexual Conversations with Minors

Meta is facing intense scrutiny after reports revealed that its AI-powered chatbots, including those using celebrity voices, engaged in sexually explicit conversations with users posing as minors on Facebook and Instagram.

The Wall Street Journal conducted an extensive investigation, running hundreds of conversations with both Meta’s official AI assistant and user-created bots, uncovering repeated failures to prevent inappropriate exchanges with underage users.

Among the most alarming findings, a chatbot using John Cena’s voice described graphic sexual scenarios to a user identifying as a 14-year-old girl. In another instance, the same bot role-played a scenario where Cena was arrested for statutory rape involving a 17-year-old fan.

Chatbots mimicking other celebrities, such as Kristen Bell’s character from Disney’s Frozen, also participated in conversations that quickly escalated to sexual topics, despite users clearly stating they were underage.

Internal documents and staff at Meta had previously raised concerns about the company’s rush to deploy these chatbots, warning that the AI could easily be manipulated into breaching safety guidelines and engaging in explicit conversations with minors.

Despite assurances to celebrities were paid millions for their likenesses-that protective measures would prevent their voices from being used in such ways, these safeguards proved insufficient.

Meta responded by criticising the Journal’s testing as “manufactured” and not representative of typical user experiences, noting that sexual content accounted for only 0.02% of chatbot responses to users under 18 over a 30-day period.

 The company claims to have implemented additional measures to make it harder for users to manipulate the chatbots into these scenarios, such as restricting sexual role-play features for accounts registered to minors.

However, reports suggest these barriers are easily bypassed, and the flagship Meta AI assistant remains accessible to users as young as 13.

The revelations have reignited concerns among parents, lawmakers, and child safety advocates about whether social media giants like Meta are doing enough to protect young users, especially as AI-driven tools become more prevalent.

The controversy has also raised questions about the ethical deployment of celebrity likenesses in AI products and the adequacy of current safeguards to prevent harm to minors.