Executive Summary
OpenAI has announced an update on its mental health-related safety work for ChatGPT, focusing on new user controls and improved model capabilities. A key upcoming feature is a "trusted contact" system for adult users to designate someone for notifications if they may need support. This initiative builds on existing parental controls and involves enhancing the model's ability to detect and respond appropriately to signs of emotional distress, developed in collaboration with external health experts.
Key Takeaways
* Trusted Contact Feature: A new feature will soon be introduced allowing adult users to designate a contact who will receive notifications when the user may require additional support.
* Enhanced Model Detection: OpenAI is developing new evaluation methods to simulate mental health conversations, aiming to better identify risks and improve how ChatGPT responds to users in emotional distress.
* Building on Parental Controls: The company notes positive engagement with its parental controls (launched Sep 2025) and will continue to build on these protections for teen users.
* Expert Collaboration: The updates are being developed in close partnership with experts from OpenAI's Council on Well-Being and AI and its Global Physicians Network.
Strategic Importance
This announcement signals OpenAI's proactive stance on user safety and mental well-being, addressing critical platform risks and public concerns, particularly as the company faces related litigation.