OpenAI's "opaque recurrence" in Astra alarms AI safety experts over CoT monitorability
dl_weekly · x · 2026-09-06
The Information reports OpenAI's new Astra model uses a "recurrent depth" reasoning technique that loops computation outside the visible chain of thought. Redwood CEO Buck Shlegeris is "extremely concerned": while Astra's use is limited, scaling recurrence could "totally destroy CoT monitorability." Zvi Mowshowitz argues laws may be needed to prevent a race to the bottom in monitorability.
Related event: OpenAI's Astra Reportedly Uses Recurrent Depth, Raising AI Safety Concerns(7 posts)→
More from Models
- ICML paper: finite context windows impose impossibility results for adversarial regret — burny_tech · 2026-11-02
- SenseTime open-sources SenseNova-Loop-9B: weight-shared 56-layer loop, 256K context, Apache 2.0 — AdinaYakup · 2026-11-02
- User says Gemini excels at long-form fanfic, producing stories up to 29 minutes of reading — Ok-Jello3739 · 2026-11-02
- Sam Altman calls tokens 'a kind of fake unit' that doesn't measure business value — rohanpaul_ai · 2026-11-02
- 'You Can Feel the Gradient Slopes': Dev Contrasts Benchmaxxed RL Models vs General Ones — theteknosaur · 2026-11-02
- LightOnOCR-3 Launches: Tops FRBench, Adds Layout Structure and Chart Extraction — IgorCarron · 2026-11-02