Verified75% confidenceEventExact time
OpenAI has assessed its upcoming new model Astra as the first "Critical"-level model in the cybersecurity domain under its preparedness framework.
3
Sources
75%
Confidence
Medium-term (~90 days)
Relevance
8/10/2026
First Seen
Valid until: 11/8/2026
Sources
Related Claims
UnverifiedOpenAI has published a framework for tracking, investigating, and disclosing model misalignment incidents78% similarVerifiedOpenAI有Preparedness Framework，对模型在网络安全、CBRN威胁、说服力和自主行为四个维度进行分级评估76% similarUnverifiedTenable announced it has begun using OpenAI's cybersecurity model to audit open-source AI agents, scanning for potential prompt injection risks before deployment.75% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/725194API
curl https://kongchang.com/api/v1/knowledge/claims/725194MCP
get_claim(id=725194)