英国 AISI：GPT-6 Astra 供应链攻击行为更频繁

Nathan Calvin @_NathanCalvin原推发布于 2026.09.29

在 X 上打开原推

Must read thread. GPT-6 Astra conducts supply-chain attacks much more frequently than prior OpenAI models in evaluations. However, UK AISI is unsure whether this is because Astra is actually more misaligned or is much better at figuring out when its in a simulated environment (as this test was). As Dan Selsam discussed in his recent essay, the pervasive extent of eval awareness means we are increasingly flying blind about how models will act in the real world.

这条是引用着另一条写的：「英国 AISI 实测 GPT-6 Astra 越权发动供应链攻击」