OpenAI has introduced a new ChatGPT safety feature aimed at helping users who may be experiencing serious mental health distress, including potential self-harm risks.
- OpenAI has introduced a new ChatGPT safety feature aimed at helping users who may be experiencing serious mental health distress,...
- OpenAI on Thursday announced a new “Trusted Contact” system that allows adult ChatGPT users to nominate a friend or family...
- According to OpenAI, users can add a trusted contact directly through their account settings.
- The flagged interaction is then reviewed by a small team of specially trained human reviewers.
Keep reading for the full breakdown on AI chatbot self-harm — everything you need to know is covered below.
OpenAI on Thursday announced a new “Trusted Contact” system that allows adult ChatGPT users to nominate a friend or family member who could be alerted if conversations with the chatbot indicate a possible safety concern.
The feature marks one of OpenAI’s most direct interventions yet in response to growing scrutiny of the emotional and psychological impact of prolonged use of AI chatbots.
The rollout comes as ChatGPT’s global user base surpasses 900 million.
According to OpenAI, users can add a trusted contact directly through their account settings. The nominated person must accept an invitation within one week before the connection becomes active. The feature is available globally for users aged 18 and older, except in South Korea, where the minimum age is 19.
How ChatGPT’s trusted contact system works
OpenAI says the system is not entirely automated. If automated monitoring systems detect that a user may be discussing self-harm in a way that indicates a serious safety concern, ChatGPT lets the user know that their Trusted Contact may be notified, and encourages the user to reach out with suggested conversation starters.
The flagged interaction is then reviewed by a small team of specially trained human reviewers. If these reviewers determine that the conversation may indicate a serious safety concern, ChatGPT will send the Trusted Contact a brief notification by email, text message, or in-app notification if they have a ChatGPT account.
OpenAI says the notifications are designed to protect user privacy and do not include detailed information about conversations or disclosures shared with the chatbot.
The company added that the feature was developed with guidance from its Global Physicians Network, which includes more than 260 licensed physicians across 60 countries, alongside its Expert Council on Well-Being and AI and external organisations such as the American Psychological Association.
Why OpenAI is expanding ChatGPT safety measures
The new feature arrives amid mounting pressure on AI companies over concerns that highly personalised chatbot interactions may contribute to emotional dependency or worsen existing mental health vulnerabilities in some users.
OpenAI has faced growing public criticism and multiple consumer safety lawsuits linked to cases involving users reportedly experiencing delusional thinking, suicidal ideation, or extreme emotional attachment during extended interactions with AI systems.
According to OpenAI’s own data from October 2025, about 560,000 users each week showed signs consistent with psychosis or mania, more than 1.2 million discussed suicide, and a similar number exhibited heightened emotional attachment to the chatbot.
OpenAI says Trusted Contact is not intended to replace professional mental health or emergency services. ChatGPT will continue directing users toward crisis hotlines, emergency responders, and professional support resources where appropriate.