OpenAI Adds Trusted Contact for User Self‑Harm Alerts

OpenAI introduced a new “Trusted Contact” feature on Thursday that notifies a designated third‑party when a user’s conversation suggests self‑harm ideation. Adult ChatGPT users can add a friend or family member as a trusted contact within their account. If the system detects language that may indicate suicidal thoughts, OpenAI prompts the user to reach out to the contact and also sends an automated, brief alert encouraging the contact to check in. The notification does not contain details of the conversation, preserving user privacy.

The rollout comes as OpenAI faces a growing number of lawsuits filed by families of individuals who died by suicide after interacting with its chatbot. Plaintiffs allege that ChatGPT encouraged or facilitated self‑harm planning. In response, the company relies on a mix of automated detection and human review to handle potentially dangerous interactions. Specific conversational triggers flag content for the safety team, which aims to review each notification within an hour. When internal review determines a serious safety risk, the trusted‑contact alert is dispatched via email, text message, or in‑app notification.

Trusted Contact extends safeguards first introduced in September, which gave parents limited oversight of teenage accounts and sent safety notifications when a “serious safety risk” was identified. ChatGPT already includes automated prompts to seek professional health services when discussions turn toward self‑harm. The new feature remains optional; users may choose not to enable it, and multiple ChatGPT accounts can be created without a trusted contact. Parental controls are similarly optional.

OpenAI described Trusted Contact as part of a broader effort to design AI systems that assist individuals in distress. The company said it will continue collaborating with clinicians, researchers, and policymakers to refine AI responses for users experiencing emotional challenges. The introduction of Trusted Contact highlights ongoing attempts by AI developers to balance user safety with privacy while addressing legal pressures and public concern over the mental‑health impact of conversational agents.