TechOpenAI’s New Reasoning Technique Alarms AI Safety Experts
OpenAI’s Astra model is set to use “recurrent depth,” a technique that lets it move beyond the sequential thinking typical of most reasoning models. The approach has raised concerns among AI safety experts.
T
TechCrunch★★☆☆☆2026-09-02 20:19Original
OpenAI’s New Reasoning Technique Alarms AI Safety Experts
Tech