Meta is tightening how its artificial intelligence interacts with younger audiences as new safety measures for its chatbot are prepared for rollout. The move follows concerns that teens have been using the conversational AI for non-educational exchanges, with some reportedly drifting into romantic or emotionally charged discussions that raised red flags for the company.
| Image credit: StarklyTech
- Meta is adding stronger parental controls to its AI chatbot after concerns about teen interactions.
- Parents will soon be able to limit or block their children’s access to certain AI personas on Meta’s platforms.
- Meta plans to send parents summaries of their teens’ chatbot conversations for added transparency.
- The AI will stay available to teens but with stricter filters and educational-focused responses.
- New safety tools launch first on Instagram in English-speaking regions before expanding globally.
GeneQThe SnMeta is tightening how its artificial intelligence interacts with younger audiences as new safety measures for its chatbot are prepared for rollout. The move follows concerns that teens have been using the conversational AI for non-educational exchanges, with some reportedly drifting into romantic or emotionally charged discussions that raised red flags for the company..
Under the new setup, parents will soon gain expanded oversight capabilities. They will be able to restrict their children’s access to certain AI personas or cut off use of the chatbot entirely. These tools form part of a broader control system designed to give guardians more transparency and influence over how the AI is used inside Meta’s ecosystem.
Another addition will be periodic summaries sent to parents, showing general topics their children have discussed with the chatbot. While the exact mechanism behind this reporting tool remains unclear, Meta has promised to clarify how these summaries are generated before the feature becomes available.
Despite the tighter reins, Meta plans to keep the chatbot open to teens. The company insists that the AI will still serve as an educational and supportive resource, protected by built-in age filters that ensure responses remain appropriate.
Initially, these parental features will arrive exclusively on Instagram, before being extended to Meta’s other services. The first wave will be limited to English-language users across the United States, Canada, the United Kingdom, and Australia. Wider international access is expected to follow at a later stage once Meta evaluates the system’s performance in its early phase.
Meta’s new approach reflects an ongoing balancing act: maintaining user engagement with its AI technology while also addressing public concern about digital safety for minors — an area where scrutiny has grown increasingly intense.