ChatGPT Expands Safety Options With New Trusted Contact Feature for Adults

OpenAI is starting to roll out Trusted Contact in ChatGPT, a new safety feature allowing adult users to name a person they trust who will be notified when the chatbot detects signs of emotional distress. This is an optional feature that mimics the functionality of the existing safety alert options parents can receive from a linked teen account.

“Trusted Contact is designed to offer another layer of support alongside the localized helplines⁠ already available in ChatGPT, by helping users connect to a person they trust when they are in crisis,” OpenAI said today. The AI startup developed this new safety feature with guidance from clinicians and safety experts, and ChatGPT users will be able to designate just one adult as their Trusted Contact, who will need to accept the invitation.

In practice, ChatGPT has monitoring systems in place to detect when a user experiences a moment of emotional crisis. When that happens, the chatbot will first tell the user that they may notify their Trusted Contact about the situation. It will also suggest them to reach out to their Trusted Contact and offer conversation starters.

From that point, a team of human reviewers will analyze the situation and decide whether to send a notification to the user’s Trusted Contact or not. “The notification is intentionally limited. It shares the general reason that self-harm came up in a potentially concerning way, and encourages the Trusted Contact to check in. It does not include chat details or transcripts to protect user privacy,” OpenAI explained.

The AI startup emphasized that while its safety systems are not perfect, “every notification undergoes trained human review before it is sent, and we strive to review these safety notifications in under one hour.” ChatGPT also has built-in safeguards to refuse harmful requests, and it’s also designed to encourage users to contact emergency services during sensitive moments.