ChatGPT Adds Trusted Contact for Crisis Support

OpenAI is rolling out a new optional feature called Trusted Contact for adult ChatGPT users. It lets you nominate a friend, family member, or caregiver who could be notified if the system detects you might be discussing self-harm in a way that signals a serious safety concern. The feature works alongside existing crisis helpline prompts within ChatGPT. When automated monitoring flags a sensitive conversation, the user is warned that their Trusted Contact may be notified and encouraged to reach out themselves with suggested conversation starters. A small team of specially trained reviewers then assesses the situation before any notification is sent. The trusted person gets an invite and must accept within a week. This builds on parental safety alerts for teen accounts, extending support to adults who choose to opt in. Why it matters: By proactively leveraging social connection — a known protective factor against suicide risk — OpenAI is embedding real-world safety nets into AI interactions, acknowledging that even the most thoughtful algorithms can’t replace human support in moments of crisis.