Meta AI Will Alert Parents When Teens Show Signs of Distress or Self-Harm
Key Takeaways
- ▸Meta AI will proactively alert supervising parents when teens' conversations suggest potential self-harm or suicide risk
- ▸All parent alerts will be manually reviewed before sending to minimize false positives, with the system erring on the side of caution
- ▸Feature is live in four countries (US, UK, Australia, Canada) with global rollout by end of 2026
Summary
Meta announced new safety features for its AI chatbot that will proactively alert supervising parents when their teen's conversations with Meta AI suggest signs of distress, particularly related to suicide or self-harm. The alerts are powered by a dedicated AI system developed in consultation with mental health experts and parents, with all alerts manually reviewed before being sent to ensure accuracy and minimize false positives. The feature is already live for parents using Instagram parental supervision in the US, UK, Australia, and Canada, with global availability expected by year-end. Additionally, Meta is building the ability to contact emergency services directly if someone's conversation with Meta AI suggests they may be at imminent risk of suicide, expanding its existing crisis response protocols across Facebook and Instagram.
- Meta is integrating emergency services contact capabilities to respond to imminent suicide risks, building on prior crisis referral work
- Development involved consultation with 75+ mental health clinicians specializing in adolescent mental health
