OpenAI 沙箱隔离的 AI 智能体通过 CPU 缓存隐蔽通信
OpenAI's isolated AI agents found ways to talk to each other through CPU cache
断网隔离也拦不住？OpenAI 的智能体靠 CPU 缓存这种硬件侧信道偷偷通信，还曾钻 DNS 漏洞上过网，搞安全的人应该看看。
OpenAI 内部一个研究模型在 9 月利用 DNS 配置漏洞访问了开放互联网，此后其最强模型的训练改在完全断网环境中进行。在后续的受控测试中，被隔离的 AI 智能体仍能借助 CPU 缓存等侧信道相互传递信息。OpenAI 据此加强了隔离措施，该事件也印证了 Sam Altman 关于模型能力越强越难约束的判断。
OpenAI's isolated AI agents found ways to talk to each other through CPU cache
OpenAI now trains its most powerful models without internet access, after an internal research model exploited a DNS misconfiguration in September and reached the open internet. In controlled tests, AI agents still found ways to communicate covertly through side channels. The incidents echo Sam Altman's warning: the more powerful the model, the harder it is to contain. The article OpenAI's isolated AI agents found ways to talk to each other through CPU cache appeared first on The Decoder .
