A story we follow
Reports of less readable reasoning in Astra raise monitoring concerns
Tracks reporting that OpenAI's Astra architecture reduces reliance on readable chain of thought, conflicting statements about current monitorability, and consequences for safety monitoring.
A story page follows one specific event across podcast discussions, with an overview and a timeline of what changed. It updates when new episodes discuss the event, and you can get those updates by email or push. How our formats work
By email or push, when we publish an update to this story. It does not subscribe you to other stories.
Overview
Reports that Astra reasons less through readable text have raised monitoring concerns. Buck Shlegeris highlighted conflicting statements from OpenAI staff about whether current models are less monitorable. Alex argued that possible depth scaling deserves more safety attention, while rejecting the idea that long-term safety must depend on readable chain of thought. OpenAI researcher Noam Brown has now said on the record that OpenAI is already seeing signs that chain-of-thought monitorability is degrading. He said models are getting better at controlling their chain of thought, and that OpenAI is still working out why so it can reverse the trend. He added that models cannot yet actually hide their reasoning. Brown did not attribute the decline to Astra's architecture. His interviewer called the Astra recurrence controversy a sideshow next to the incident record. The architectural premise itself remains unconfirmed.
What changed
Dates show when each podcast discussion was published.
-
New information
On the Dwarkesh Podcast, OpenAI researcher Noam Brown said OpenAI is already seeing signs that chain-of-thought monitorability is degrading. He said models are getting better at controlling their chain of thought and that OpenAI is investigating why, because it wants to reverse the trend. In his account, models know about chain-of-thought monitoring from pre-training data but cannot yet actually hide their reasoning. If they could, OpenAI would have to rely on other ways of seeing what models do. He did not tie the decline to Astra's reported architecture. The interviewer called the controversy over Astra's recurrent or looped transformer a sideshow compared with chain of thought failing to prevent the Hugging Face incident.
-
Interpretation
Alex frames reduced reasoning visibility as the substantive concern beneath kill-switch publicity, but treats depth scaling as a hypothesis and questions permanent reliance on token-level oversight.
-
Interpretation
Shlegeris highlights the safety cost of reported changes to Astra's reasoning architecture while explicitly acknowledging that OpenAI staff statements leave current monitorability uncertain.
Podcast discussions
- Dwarkesh Podcast Noam Brown – Agent swarms, alignment, & recursive self-improvementSep 17, 2026
- Moonshots with Peter Diamandis GPT-6 Astra Saturates ARC-AGI-3, Tesla's $30K Cybercab Floods Austin, Anthropic Proves Fermat's Last Theorem | EP #286Sep 5, 2026
- Unsupervised Learning Ep 93: CEO of Redwood Research Buck Shlegeris on OpenAI/HuggingFace Revelations, Fixing AI Safety & Takeover OddsSep 3, 2026
Sources
- 01
- 02
- 03
Version history
-
Sep 29, 2026 · Version 3
On the Dwarkesh Podcast, OpenAI researcher Noam Brown said OpenAI is already seeing signs that chain-of-thought monitorability is degrading. He said models are getting better at controlling their chain of thought and that OpenAI is investigating why, because it wants to reverse the trend. In his account, models know about chain-of-thought monitoring from pre-training data but cannot yet actually hide their reasoning. If they could, OpenAI would have to rely on other ways of seeing what models do. He did not tie the decline to Astra's reported architecture. The interviewer called the controversy over Astra's recurrent or looped transformer a sideshow compared with chain of thought failing to prevent the Hugging Face incident.
-
Sep 15, 2026 · Version 2
Alex frames reduced reasoning visibility as the substantive concern beneath kill-switch publicity, but treats depth scaling as a hypothesis and questions permanent reliance on token-level oversight.