OpenAI News·· 8 d ago
Introducing MentalHealthBench
Introducing MentalHealthBench
AI summary
OpenAI released MentalHealthBench, built with experts to assess whether AI responses in real mental health conversations are helpful and safe.
Selection record
Threshold 60Official, first-handFirst 32Second 27
Not admittedSum of both 59 < twice the threshold 120
- Source tier
- Official, first-hand; this tier's threshold is 60
- Pre-filter
- passed：OpenAI发布AI心理健康评测基准
A model scores each item twice, independently, against one written standard, out of 100. An item is admitted only when the two scores add up to twice the threshold. The threshold is set per source tier.
Source: OpenAI News · openai.com