Critic: OpenAI called for oversight only after its CoT-monitorability architecture decision
scaling01 · x · 2026-09-02
AI safety commentator scaling01 criticizes OpenAI for publicly endorsing oversight only after making an architecture decision affecting chain-of-thought monitorability, despite having argued that lab architectural decisions should be scrutinized by independent safety-minded experts through legislated audits.
Core accusations:
- OpenAI called for oversight after the fact and invited no safety researchers or third parties to review the decision beforehand
- It released no study or further information on how the decision affects CoT monitorability
- The author demands OpenAI publish its research on CoT monitorability and explain how it intends to monitor models going forward
Related event: Critics urge OpenAI to adopt industry-wide AI monitorability standards(4 posts)→
More from Safety
- One diagram explains the difference between CSRF and XSS attacks — _jaydeepkarale · 2026-10-18
- Grady Booch slams Anthropic for tolerating up to 50 deaths and $1B damages per model — Grady_Booch · 2026-10-18
- Single Neuron Suffices to Bypass Safety Alignment in LLMs, Paper Finds — burny_tech · 2026-10-18
- Polymarket gives OpenAI just 8% odds of a full training pause by end of 2026 — Polymarket · 2026-10-18
- OpenAI engineer over alleged hidden recording device seeks bail, denies Saudi escape plan — Polymarket · 2026-10-18
- Free comic "The Collective" offers a strange retelling of the Hugging Face hack — Flugelwagen · 2026-10-18