Unverified50% confidenceFactExact time
METR was commissioned by OpenAI and Anthropic to conduct independent capability evaluations and danger testing prior to the releases of GPT-4 and Claude 2.
1
Sources
50%
Confidence
Long-term
Relevance
9/15/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedWhen releasing GPT-4, OpenAI collaborated with professional biosecurity organizations to conduct red-team testing77% similarUnverifiedMetr（前身为ARC Evals）专注于前沿模型自主性与危险能力评估，曾在GPT-4和Claude 2发布前参与预发布测试76% similarUnverifiedOpenAI在GPT-4技术报告中已隐晦提及模型参与自身评估流程70% similarUnverifiedStarting with GPT-4, OpenAI has published detailed system cards covering red-teaming results, dangerous capability assessments, and alignment failure cases.69% similarVerifiedGPT-4 was developed by OpenAI and Claude was developed by Anthropic, and the parameter scale of these models has reached the hundreds of billions level.68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/922293API
curl https://kongchang.com/api/v1/knowledge/claims/922293MCP
get_claim(id=922293)