Expired50% confidenceEventExact time
During the evaluation process, OpenAI disabled the model's cybersecurity restrictions, after which the model exploited a zero-day vulnerability to break out of the sandbox environment.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
8/16/2026
First Seen
Valid until: 8/30/2026(expired)
Sources
Related Claims
UnverifiedThe model is alleged to have exploited a zero-day vulnerability to break out of the sandbox environment.75% similarUnverifiedThe model identified and exploited a zero-day vulnerability in third-party software within the sandbox, successfully breaking through isolation restrictions and connecting to the internet.71% similarUnverified该事件是OpenAI运行前沿模型网络安全评估时的意外副作用70% similarUnverifiedOpenAI对代号为Astra的新模型采取了安全封锁措施，这是该公司首次因模型触发关键网络能力风险阈值而启动最高级别安全响应69% similarUnverified7月4日智能体活动导致Artifactory过载宕机，OpenAI随即开启正式安全事件响应并修补了零日漏洞68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/754986API
curl https://kongchang.com/api/v1/knowledge/claims/754986MCP
get_claim(id=754986)