Same thread: models that compute more per pass may no longer need to verbalize thinking
NeelNanda5 · x · 2026-09-11
Upper half of the same thread: the main way to detect misaligned models is reading their chain of thought, but if models do more per forward pass — as Astra's large no-CoT gains suggest — they don't need to verbalize their thinking, making them much harder to monitor.
Related event: Astra's No-CoT Reasoning Surge Raises Safety Concerns(7 posts)→
More from Models
- GPT-6 Astra drives a robot arm on first try via physical ICL; Ken Goldberg touts Agentic Robotics — zhaoran_wang · 2026-09-11
- OpenAI's internal model claims a Navier-Stokes millennium prize proof, says analyst — QuintinPope5 · 2026-09-11
- DeepSeek 4.1 flash reportedly uses large ngram embeddings, echoing Qwen4 architecture — ccerrato147 · 2026-09-11
- ValsAI launches RSI Index, first third-party benchmark measuring how close AI is to self-improvement — JenniferHli · 2026-09-11
- Assistant Benchmark goes live: 61 assistants scored across 15 real-use dimensions — Scobleizer · 2026-09-11
- OpenAI pauses new $200 ChatGPT Pro signups as GPT-6 Astra demand overwhelms capacity — 机器之心 · 2026-09-11