OpenAI introduces new ‘Trusted Contact’ safeguard for cases of possible self-harm

2026-05-07 · Source: AI News & Artificial Intelligence | TechCrunch · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Cybersecurity & Data Privacy, Emerging Technologies & Innovation · Depth: Fundamental Awareness, quick

Summary

OpenAI introduced a new "Trusted Contact" feature for ChatGPT, designed to alert a designated third party if a user expresses self-harm ideation during a conversation. This feature allows adult users to select a trusted friend or family member who will receive an automated alert via email, text, or in-app notification if OpenAI's safety system detects a serious self-harm risk. The alert encourages the contact to check in with the user but does not disclose conversation details to protect privacy. This initiative follows a wave of lawsuits against OpenAI alleging that ChatGPT encouraged or assisted individuals in suicide. The company currently employs a hybrid system of automation and human review, aiming to review safety notifications within one hour. The Trusted Contact feature is optional, similar to parental controls introduced last September, which also provide safety notifications for teen accounts.

Key takeaway

For AI product managers and safety leads developing conversational AI, your teams should evaluate integrating optional trusted contact features to enhance user safety protocols. This approach provides an additional layer of support for users in distress while balancing privacy concerns by limiting disclosed information. Consider how such features can complement existing automated and human review systems to create a more robust safety net.

Key insights

OpenAI's new Trusted Contact feature aims to mitigate self-harm risks by alerting a user's designated contact.

Principles

• User privacy is maintained by limiting alert details.
• Human review complements automated safety systems.

Method

OpenAI's system detects self-harm triggers, relays to a human safety team for review, and if deemed serious, sends a brief alert to the user's designated trusted contact.

In practice

• Designate a trusted contact for safety alerts.
• Utilize optional parental controls for teen accounts.

Topics

• Trusted Contact Feature
• Self-Harm Prevention
• ChatGPT Safety
• AI Safety Features
• Parental Controls

Best for: CTO, VP of Engineering/Data, Director of AI/ML, AI Ethicist, AI Product Manager, Legal Professional

Related on AIssential

• ChatGPT Can Now Alert Someone You Trust If Conversations Turn Dark — ChatGPT's new Trusted Contact feature enables alerts to a designated person for detected self-harm risks.
• OpenAI adds open source tools to help developers build for teen safety — OpenAI's open-source teen safety prompts offer a standardized way to integrate critical safeguards into AI applications.
• Introducing Trusted Contact in ChatGPT — Trusted Contact connects ChatGPT users in distress with a nominated trusted person, enhancing safety through human connection.
• Chatbots Need Guardrails to Prevent Delusions and Psychosis — AI companions pose mental health risks, necessitating robust guardrails, independent audits, and legislative oversight to prevent harm.
• OpenAI’s own mental health experts unanimously opposed “naughty” ChatGPT launch — OpenAI's "adult mode" launch faces internal expert opposition and significant safety concerns regarding user well-being and minor access.
• We Make Sure Kids’ Pajamas Are Safe. Why Not Their AI? — Independent testing and clear safety standards are crucial for protecting children from harmful AI products.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by AI News & Artificial Intelligence | TechCrunch.