0
Applied AI·September 2, 2026·1 min read

OpenAI’s new reasoning technique alarms AI safety experts

Share

Recurrent depth in Astra — reasoning outside fixed sequential passes — is a qualitative shift in how models think, which is why safety experts are nervous. If you’re considering next-gen reasoning models, budget extra time for evals and red-teaming; you can’t assume your current test harness captures new failure modes.