ChatGPT adds trusted contact support for serious safety moments

OpenAI’s ChatGPT release notes say trusted contact is rolling out from May 7, 2026 over the coming weeks for eligible adults with personal ChatGPT accounts.

The feature lets a user choose someone they trust in settings. If automated systems and trained reviewers detect a serious safety concern around suicide, ChatGPT may notify the trusted contact and encourage them to check in.

OpenAI says the feature is available only for eligible personal ChatGPT accounts in supported regions. It is not available for Business, Enterprise, or Edu workspaces.

Why this matters

General-purpose AI assistants are increasingly used for emotional support, planning, health questions, and private conversation. That makes crisis-path design part of product quality, not a side policy page.

Trusted contact also changes the privacy expectation for personal ChatGPT. Users who enable it are explicitly creating a potential human escalation path for a narrow class of safety events.

Buyer take

For personal users, trusted contact may be valuable if ChatGPT is part of your support routine, but it should not replace local emergency services, clinicians, or trusted people in your life.

For organizations, note the boundary: this is not a Business, Enterprise, or Edu control. Workspace admins should not assume trusted contact exists as a managed workplace safety feature.

What is still unclear

OpenAI’s release note does not list supported regions, the full eligibility rules, or the exact review and notification thresholds. Users should read the dedicated support documentation before enabling it.