Chinese Eval Startup Debates CoT Monitorability with Kokotajlo

A Chinese AI safety evaluation startup engaged in a public exchange with ex-OpenAI researcher Daniel Kokotajlo over CoT monitorability evaluation methods. Kokotajlo raised a key objection: what if the model doesn't refuse to corrupt its own chain of thought.

2026-10-27 ~ 2026-10-27 · 2 related posts

• Kokotajlo challenges CoT monitorability evals: what if the model didn't refuse? — dfrsrchtwts · 2026-10-27
• Daniel Kokotajlo debates a Chinese AI safety startup on whether CoT monitorability evals prove enough — S_OhEigeartaigh · 2026-10-27