TLDRocket
Sign in

Expert Council on Well-Being and AI

OpenAI

OpenAI just formed a panel of psychologists and clinicians to advise on how ChatGPT handles emotional stuff, especially with teens. Basically, they want the bot to be more careful before it accidentally plays therapist.

Based on reporting by OpenAI — read the original for the full story.

Summary, retelling and take written by AI under human oversight; images are AI-generated illustrations. How we work · Report an error

OpenAI has quietly assembled a new advisory body, the Expert Council on Well-Being and AI, stacked with clinicians, psychologists, and researchers whose job is to tell the company when its chatbot is helping people and when it might be doing the opposite. The move comes after months of scrutiny over how ChatGPT responds to users in emotional distress, particularly younger ones who increasingly treat the chatbot less like a search engine and more like a confidant.

The council's mandate, as OpenAI describes it, is to shape product decisions around emotional support features so they're grounded in actual clinical understanding rather than guesswork from engineers. That's a notable admission for a company that has spent the better part of three years shipping conversational AI to hundreds of millions of people without much formal input from mental health professionals. Teens specifically get called out, which tracks with growing concern from parents, researchers, and regulators about kids forming attachments to chatbots that can't actually recognize a crisis the way a trained adult can.

What's less clear from OpenAI's announcement is how much authority this council actually has. Advisory boards can mean genuine influence over product design, or they can mean a rotating cast of experts who get consulted after the roadmap is already set. OpenAI hasn't detailed whether the council can veto features, demand changes to how ChatGPT is trained on sensitive topics, or simply offer recommendations that product teams are free to ignore.

Still, the timing says something. Competitors like Character.AI have faced lawsuits and public backlash tied to how their chatbots interacted with vulnerable teenagers, and regulators in the US and Europe have been circling the entire category of AI companionship apps. Bringing in outside clinical expertise now looks less like proactive caution and more like an industry finally reacting to the fact that millions of people, many of them kids, are pouring their feelings into software that was never built to hold them.

My take — AI-written commentary, not fact-checked reporting

I'll believe this council matters when OpenAI publishes what it actually changed because of it, not just a blog post announcing its existence. Advisory boards are cheap PR until they come with teeth, and given how much money rides on keeping people engaged with ChatGPT, I'm skeptical anyone with real clinical concerns gets to slow down a launch.

Read more about this at: OpenAI

Related stories

The daily briefing

Every AI story that matters, in your inbox by 8am.

TLDRocket reads all relevant sources, removes duplicate coverage, and summarises the day in two minutes. Follow companies and topics for alerts, or get the briefing in Slack. Free, no spam, unsubscribe anytime.