What is ChatGPT's new mental health safety feature?

ChatGPT has launched a "trusted contact" feature for adult users, allowing them to designate a contact who may be notified if the AI detects discussions about self-harm.

How does the trusted contact feature work?

If ChatGPT's systems detect concerning discussions about self-harm, the user will be encouraged to reach out to their trusted contact, and trained reviewers will assess if a notification should be sent.

Does this feature replace professional mental health care?

No, OpenAI states that the trusted contact feature is designed to encourage connection and support and does not replace professional care or crisis services.

Home / Technology / ChatGPT Adds Trusted Contact for Mental Health Support

ChatGPT Adds Trusted Contact for Mental Health Support

8 May

•

Summary

Adult ChatGPT users can designate a trusted contact for self-harm alerts.
The feature alerts a chosen contact if AI detects concerning self-harm discussions.
This aims to encourage connection and support for users in distress.

ChatGPT Adds Trusted Contact for Mental Health Support

OpenAI has introduced a new mental health safety feature for adult ChatGPT users. The "trusted contact" system allows users to designate an adult who may be notified if the AI detects discussions about self-harm.

This feature, developed with guidance from mental health experts, aims to foster connections with people users already trust. It's designed as an additional layer of support, not a substitute for professional help or crisis services.

When the AI flags potential self-harm discussions, users will be prompted to reach out to their designated contact. Trained reviewers then assess the situation before deciding to send an alert. Notifications will protect user privacy by omitting specific chat details.

Disclaimer: This story has been auto-aggregated and auto-summarised by a computer program. This story has not been edited or created by the Feedzop team.