Meta AI alerts parents when teens discuss self-harm with chatbot

Meta announced a new AI-powered parental alert system that monitors conversations between teens and Meta's chatbot for signs of self-harm or suicidal ideation. When concerning language is detected, the supervising parent receives an immediate notification through their Meta account, enabling early intervention. The feature builds on existing safeguards that already direct teens to crisis helplines and encourage them to talk to trusted adults. Meta also plans to expand capabilities to contact emergency services if a conversation indicates imminent risk, complementing the 19,000+ referrals it made globally last year based on posts.

The tool is currently live for Instagram parental supervision users in the US, UK, Australia, and Canada, with a global rollout planned by year-end. Additionally, Meta is enforcing stricter content settings for teen accounts: all teen accounts default to a 13+ content filter that also applies to Meta AI interactions, preventing the chatbot from complying with sensitive prompts while providing age-appropriate responses. Meta emphasized that the feature respects privacy safeguards while giving families timely awareness to ensure teens receive the offline support they need, whether from parents, professionals, or emergency services.