TechCrunch · AI· Aditya Mehta·· 14 h ago
Anthropic is letting METR researchers look under the hood with employee-level access
Anthropic is letting METR researchers look under the hood with employee-level access
AI summary
METR will publish its own assessments of Anthropic’s risks, incidents, and practices, and the frontier AI model maker has given up the right to block them.
Selection record
Threshold 76Media and individualsFirst 71Second 62
Not admittedSum of both 133 < twice the threshold 152
- Source tier
- Media and individuals; this tier's threshold is 76
- Pre-filter
- passed：Anthropic开放前沿AI模型独立评估
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: TechCrunch · AI · techcrunch.com