Meta alerts parents when teens discuss suicide with AI

Meta announced on July 16, 2026, that it will now proactively alert parents when their teenagers discuss suicide or self-harm with the company's Meta AI chatbot on its platforms. The new safety feature, which is currently available in the United States, Canada, the United Kingdom, and Australia, requires parents to have previously opted into Meta's parental supervision tools for Instagram.

When a teen's conversation with Meta AI suggests they may be considering self-harm, the system already directs the user to crisis helplines and encourages them to speak with a trusted adult. Under the new policy, an alert will also be sent to the supervising parent. The company stated it developed a dedicated AI system to identify such conversations, aiming to detect even subtle references. However, all chats flagged by the AI will undergo a manual review by a human before any notification is sent to a parent.

Officials said they will err on the side of caution, meaning parents may sometimes be alerted even when a teen's intent is ambiguous. Along with the alert, Meta will provide parents with expert resources to help them approach these difficult conversations with their children. The feature...