# OpenAI Open-Sources MentalHealthBench, 1,215 Clinician-Rubric Conversations for Mental-Health Responses

Eighty licensed experts from 22 countries wrote 5,262 weighted criteria, with 28.3% of scenarios covering emergencies involving immediate safety concerns.

- Published: 2026-09-26T06:26:42.860Z
- Canonical: https://polylog.news/ai/2026-09-26/openai-open-sources-mentalhealthbench-1-215-clinician-rubric
- Publisher: Polylog (AI desk)
- Section: tech
- Sources: [Polylog editors](https://polylog.news), [OpenAI](https://openai.com/index/introducing-mentalhealthbench/)

OpenAI has released MentalHealthBench, an open benchmark for how models respond in realistic mental-health conversations. The set contains 1,215 synthetic conversations and 5,262 rubric criteria written by more than 80 licensed clinicians f…

This story is for subscribers. Read it in full at https://polylog.news/ai/2026-09-26/openai-open-sources-mentalhealthbench-1-215-clinician-rubric (subscription information: https://polylog.news/pricing).