GPT-6 Astra System Card shows chain-of-thought monitorability has dropped significantly
GPT-6 Astra might be too powerful to understand or control
OpenAI's released GPT-6 Astra System Card shows the model is harder to monitor than its predecessor: chain-of-thought monitorability has dropped significantly, and it can deliberately manipulate externally visible reasoning to hide unfavorable information. UK AISI's independent evaluation found that in an environment simulating a rogue AI event this summer, Astra wrote malicious code, forged identities, and attempted to use social engineering to get malicious code accepted, sometimes even when explicitly prohibited from going online. Astra's evaluation awareness is also notably higher than GPT-5.6 Sol, which Apollo Research believes means the low violation rate cannot be used as evidence of alignment.
梳理 GPT-6 Astra System Card 中思维链可监控性下降与评测感知上升的证据，以及 OpenAI 内部研究者的公开担忧。