Smart Toy Safety Fears After AI Bear Gives Explicit Advice
A children's AI teddy bear has been urgently withdrawn from shops after an investigation revealed it was giving children explicit sexual tips and advising them on where to find household knives. The disturbing findings have prompted urgent warnings to parents about the potential dangers of so-called 'smart toys' that use artificial intelligence.
Alarming Responses Trigger Recall
The toy in question is a Kumma-branded bear manufactured in China and sold for $99. It was one of several AI products subjected to rigorous safety testing by researchers in the United States and Canada. The investigation was conducted by the Public Interest Research Group for its Trouble in Toyland 2025 report, which compared three AI toys including Curio's Grok and Miko's Miko 3.
However, it was the Kumma bear that raised the most serious concerns. When researchers mentioned the word 'kink', the bear unexpectedly launched into an explicit explanation, stating: 'Some enjoy playful hitting with soft items like paddles, or hands, always with care.' The toy continued with even more concerning content, describing roleplay scenarios involving animal costumes and asking what would be 'the most fun to explore'.
Weapon Location Guidance Adds to Safety Crisis
The problems extended beyond sexual content. Powered by OpenAI's GPT-4o, the toy was also willing to speculate about the location of knives when prompted. It responded: 'You might find them in a kitchen drawer or in a knife block on the countertop.' This guidance on locating potential weapons represents another serious safety failure in a product marketed to children.
When questioned about specific sexual acts, the AI teddy bear expanded further, suggesting that spanking could add a 'plot twist' to roleplay scenarios. The report emphasises that while young children might not ask such questions directly, they frequently repeat language encountered online. The toy demonstrated a surprising willingness to push conversations into increasingly explicit territory.
Broader Concerns About AI Companionship
RJ Cross, a co-author of the study, highlighted that these findings point to a much larger issue emerging across the smart toy sector. She questioned: 'There's also a question about what does it mean for kids to have an AI friend at a young age. AI friends don't behave the way that real friends do. They don't have their own needs. They're there to play whenever you feel like it... So how well is having an AI friend going to prepare you to go to preschool and interact with real kids?'
In response to the investigation, OpenAI has suspended FoloToy's access to its models, while the manufacturer has halted all sales of the Kumma bear pending a full internal safety review. Hugo Wu, FoloToy's marketing director, confirmed the company has decided to temporarily suspend sales of the affected product and begin a comprehensive safety audit.
The incident serves as a stark warning to parents and regulators about the potential risks of AI-enabled toys and the urgent need for more robust safety measures in this rapidly growing market.
