Industry change
OpenAI releases MentalHealthBench
- Published
- Source
- OpenAI Blog
Summary
OpenAI introduced MentalHealthBench, an open benchmark for measuring how AI systems respond in realistic mental health conversations. It was co-created with more than 80 licensed mental health experts from 22 countries and evaluates responses across everyday, high-acuity, and emergency scenarios, weighing safety, context-seeking, and respect for user agency. OpenAI states clearly that ChatGPT is not a substitute for therapy or professional care.
Why it matters for our work
As huge numbers of people bring well-being questions to AI, expert-written rubrics now make good responses openly testable. It lays groundwork for collaboration between human experts and AI in mental health, though real-world impact should follow external validation.
Translated from the Korean original. Summaries may be translated and edited. Commentary reflects our perspective; forecasts remain the source’s views.