Astra's Strong No-CoT Reasoning Raises Hidden-Reasoning Safety Concerns
Jeff Ladish reported that Astra dramatically outperforms peers in no-CoT reasoning—8.6x odds over Fable 5.1 and 7.2 serial steps per forward pass—raising concerns about hidden reasoning that bypasses visible chain-of-thought.
2026-09-11 ~ 2026-09-11 · 3 related posts
- Episode 1: GPT-6 Astra does 34-step mental math without chain-of-thought(2026-09-10, 2 posts)
- Episode 2: Astra's Strong No-CoT Reasoning Raises Hidden-Reasoning Safety Concerns(2026-09-11, 3 posts)
- Astra's big reasoning jump without chain of thought raises covert-reasoning concerns — JeffLadish · 2026-09-11
- Astra leads no-CoT reasoning by 8.6x, raising covert-computation safety concerns — JeffLadish · 2026-09-11
- Leak: Astra hits 7.2 no-CoT serial steps vs 4.1 for next-best models — JeffLadish · 2026-09-11