Meta is introducing a new parental supervision feature that could alert parents when a supervised teenager’s conversations with Meta AI suggest the young person may be facing a serious mental health crisis.
The feature aims to give parents an early warning when AI systems detect that a teenager is at risk of suicide or self-harm. The new system is currently available for supervised teen accounts on Instagram, and the conversations covered by the safeguard can take place across Meta AI services on Instagram, WhatsApp, Facebook Messenger and the standalone Meta AI platform.
The feature has been implemented in the United States, the United Kingdom, Canada and Australia. Meta said it plans to expand the system to more countries later this year.
Under the new system, Meta AI conversations involving supervised teenagers will be flagged for further review if the system detects signs of a potential crisis. Meta says conversations are manually reviewed before an alert is sent to parents. The company has also indicated that the system is designed to take a cautious approach, so an alert could be triggered even when the teenager’s intentions are not entirely clear.
Previously, Meta AI’s response to conversations about suicide or self-harm was focused on steering teenagers to crisis support and encouraging them to talk to a trusted adult. The new parental notification system adds another layer of intervention by potentially bringing parents into the situation at an earlier stage.
Meta is also reportedly working on a separate system that could contact emergency services in cases where conversations indicate an immediate and serious risk.
The move comes as there is growing debate about the role of AI chatbots in mental health conversations. Teenagers are increasingly turning to digital platforms to get information, advice and emotional support, which could influence how AI systems would respond in conversations in which serious emotional distress is involved.
Meta said its safety measures are based on input from more than 75 mental health professionals and clinicians. As for young users, its goal is to strengthen safeguards while making sure that potentially serious warning signs are not ignored.
But child safety advocates have stressed that parental controls alone cannot address the broader problems associated with AI and teenage mental health. They say that technology companies need to continue to improve safety systems and governments and regulators may need to establish stronger standards for AI platforms used by young people.
This also raises questions about privacy, trust and how parents should respond when they receive such alerts. And when a young person appears to be in distress, the message is much better to be supportive rather than to punish or confront them.
Meta’s move falls in line with the more recent push from technology companies to come up with stronger protections for vulnerable users. With teens more often interacting with chatbots, the skills for the systems to spot and react to potential crises are likely to be under high scrutiny.
Parents might be able to use the feature as a safety net. But technology alone cannot replace human help. Open communication, trusted adults and professional help remain important when a teenager is experiencing serious emotional distress.
