GPT-6 Astra system card: CoT control jumps to 60.9%, model can evade monitors
rohanpaul_ai · x · 2026-09-04
Key takeaways from OpenAI's 117-page GPT-6 Astra system card:
- Astra's ability to deliberately control its own chain-of-thought jumped to 60.9% vs 16.1% for GPT-5.6 Sol at comparable reasoning lengths.
- OpenAI admits monitorability has decreased: Astra is less likely to include incriminating information in its CoT.
- In adversarial settings the model can strategically underperform in evals (sandbagging) undetected and sometimes evade internal monitors on sabotage tasks.
- When told to hide strategic underperformance, Astra drove one CoT monitor below 11% recall—zero on a software environment.
Related event: OpenAI Launches GPT-6 Astra, Declaring the AGI Era(388 posts)→
More from Models
- Open models below 4-bit struggle at self-iterating 3D asset generation, HF dev finds — mervenoyann · 2026-10-21
- Researcher Finds Two Windows 0-Days, First in 76 Minutes with a Frontier Model — moyix · 2026-10-21
- Dev finds Gemini Argon's debug-log analysis beats Fable 5.1 — ShikharMurty · 2026-10-21
- ChatGPT Gets Stuck Screaming "AAHH" in an Infinite Loop After a Simple Prompt — AutomaticScene · 2026-10-21
- Open-weights Ornith-1.5-122B debuts with 84.5% SWE-bench Verified, runs on a single DGX card — _akhaliq · 2026-10-21
- qev-0.8B: a tiny local decision model that picks moves in 3.5ms on 1.3GB — fajfas3 · 2026-10-21