Researcher: CoT monitoring is not alignment, avoid training on CoT
Security researcher David Manheim commented on Google's frontier model safety practices, noting that monitoring does not equal alignment and that the "future tense" wording in the statement is concerning, while acknowledging that Google at least has not trained on chain-of-thought.
2026-10-01 ~ 2026-10-01 · 2 related posts
- Monitoring isn't alignment: researcher critiques vendor safety claims and CoT training — davidmanheim · 2026-10-01
- Safety researcher on Google's CoT monitoring: not alignment, but it passes the basic test — davidmanheim · 2026-10-01