OpenAI News·· 19 h ago
Our first phase of embedded assessments
Our first phase of embedded assessments
AI summary
OpenAI begins embedded assessments with Apollo Research, METR, and Transluce, and Goodfire to independently evaluate model alignment, safeguards, and monitoring systems.
Selection record
Threshold 60Official, first-handFirst 42Second 58
Not admittedSum of both 100 < twice the threshold 120
- Source tier
- Official, first-hand; this tier's threshold is 60
- Pre-filter
- passed:OpenAI模型对齐与安全评估
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