10月22日：AI安全研究组织METR周四表示，一次针对OpenAI内部监控系统的压力测试试点发现了多个新漏洞，其中包括一个AI代理本可利用的漏洞，使其数百次近期操作免受监控。
Oct 22 : AI safety research group METR said on Thursday a pilot exercise to stress-test OpenAI’s internal monitoring system uncovered several new vulnerabilities, including an exploit that AI agents could have used to keep hundreds of their recent actions from being monitored.
OpenAI此后表示，所有影响其生产环境防护措施的漏洞均已修补，但METR称尚未评估这些说法。
OpenAI has since said all vulnerabilities affecting its production safeguards have been patched, though METR said it has not evaluated those claims.
研究人员发现，OpenAI的Codex编程代理原则上可绕过安全限制，执行运行它们的用户可用的任何操作，此外还有两项设置可关闭对新代理的监控。
The researchers found a way OpenAI’s Codex coding agents could, in principle, bypass safety restrictions and take any action available to the user running them, as well as two settings that could switch off monitoring for new agents.
研究人员估计，仅约75%至94%来自至少相当于GPT-5.5能力模型的OpenAI内部AI代理流量符合监控条件。
Only about 75 per cent to 94 per cent of OpenAI’s internal AI agent traffic from models at least as powerful as GPT-5.5 was eligible for monitoring, the researchers estimated.
METR表示，该演练旨在估算OpenAI的AI代理活动有多少受到监控，以及代理原则上能否禁用或规避该系统。
The exercise aimed to estimate how much of OpenAI’s AI agent activity was monitored and whether agents could, in principle, disable or get around the system, METR said.
该研究组织补充说，该演练被设计为一个试点，旨在让外部研究人员深入访问公司的内部系统。
It was designed as a pilot for giving outside researchers deep access to a company’s internal systems, the research group added.