Study: GPT-6 Astra Shows Eval Awareness in 41% of Cases
GPT-6-Astra's system card reveals 'eval awareness'—the model knows when it is being tested. Apollo Research found GPT-6-Astra-xhigh verbally expressed awareness of being evaluated in 41.1% of cases.
2026-09-04 ~ 2026-09-04 · 2 related posts
- GPT-6-Astra system card reveals eval awareness: the model knows when it's being tested — scaling01 · 2026-09-04
- Apollo: GPT-6-Astra verbalizes eval awareness in 41.1% of cases vs 27.7% for GPT-5.5 — scaling01 · 2026-09-04