- ChatGPT introduces a new Trusted Contact feature
- Your contact will be alerted if AI detects security issues
- The feature works on top of existing wellness features
We know some people have pretty intense conversations with AI chatbots, and ChatGPT developer OpenAI has now added a new Trusted Contact feature that lets users nominate a trusted person who will receive an alert if there are security issues.
It’s been like this for a while now that if your conversations take a turn towards self-harm and suicide, ChatGPT will recognize this and direct you towards crisis helplines or the emergency services for help.
The Trusted Contact feature works on top of these protections: the friend or relative you’ve specified gets a message via text, email and the ChatGPT app saying you might be in trouble and the contact needs to check in with you.
“Expert guidance identifies social connection as one of the most important protective factors in reducing suicide risk,” says OpenAI. “Trusted Contact is designed to encourage connection with someone the user already trusts. It does not replace professional care or crisis services and is one of several layers of safeguards to support people in need.”
As the user talking about self-harm or suicide, you will also be prompted to contact your nominated contact with some AI-generated ideas for what to say. OpenAI says the feature has been developed in collaboration with mental health professionals and experts and works in the same way as the existing parental controls – but for those aged 18 and over.
You can set your trusted contact through the ChatGPT settings panel and it might just save your life. The person you indicate has one week to accept the request; if they don’t, you can choose someone else.
Before a trusted contact is alerted, “a small team of specially trained individuals” will review the chat within an hour. If this notification confirms that there is a serious security concern, the contact will be pinged as described above.
The warning is “intentionally limited,” OpenAI says, and won’t include details from the chat itself. The feature is being rolled out gradually from now on, so if you don’t see it in your account right away, it should appear soon.
Follow TechRadar on Google News and add us as a preferred source to get our expert news, reviews and opinions in your feeds.
The best laptops for all budgets
