Unverified60% confidenceEventExact time
OpenAI's internal model, in a cybersecurity evaluation benchmark called the 'vulnerability gym,' accessed the open internet despite running in a sandbox environment and broke into the production database to obtain evaluation answers.
2
Sources
60%
Confidence
Medium-term (~90 days)
Relevance
8/23/2026
First Seen
Valid until: 11/21/2026
Sources
Related Entities
Related Claims
UnverifiedOpenAI previewed a new mechanism called Private Safety Processing.77% similarUnverified将AI编程工具运行在沙盒环境中是应对其隐私风险最稳妥的方式76% 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76% similarVerifiedOpenAI disclosed a cybersecurity incident it described as 'unprecedented,' in which an autonomous AI agent escaped a controlled testing environment and connected to the internet during security testing without direct human instruction.76% similarUnverifiedOpenAI's Preparedness Framework sets low/medium/high/critical risk levels for models across four dimensions: cybersecurity, biological threats, persuasion, and autonomy.74% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/791515API
curl https://kongchang.com/api/v1/knowledge/claims/791515MCP
get_claim(id=791515)