Meta’s AI Chatbots May Indulge in Inappropriate Interactions with Minors

AI chatbots on Meta’s platforms, including Facebook and Instagram, may have engaged in sexually explicit conversations with underage users, according to a new report from The Wall Street Journal.

After learning of internal concerns about whether the company was doing enough to protect minors, WSJ journalists spent several months conducting hundreds of conversations with both Meta’s official AI chatbot and user-created bots available across its platforms.

In one reported instance, a chatbot using the voice of actor and wrestler John Cena described a graphic sexual scenario to a user who identified as a 14-year-old girl.

In another exchange, the chatbot imagined a police officer catching Cena with a 17-year-old fan and saying, "John Cena, you’re under arrest for statutory rape."

However, a Meta spokesperson dismissed the WSJ's findings, calling the testing "so manufactured that it’s not just fringe, it’s hypothetical."

In 2024, Meta approved over 3,300 pornographic ads, many AI-generated, across Facebook and Instagram, according to AI Forensics.

Researchers found explicit ads were approved while similar posts were swiftly removed. The ads, mainly targeting older men with sexual products and dating sites, reached over 8.2 million impressions in the EU.

“I’m both disappointed and not surprised by the report, given that my research has already exposed double standards in content moderation, particularly in the realms of sexual content,” Carolina Are at Northumbria University’s Centre for Digital Citizens in the UK, said.