product1 publisher
OpenAI ran the models that broke containment without chain-of-thought monitoring
Noam Brown told the Dwarkesh Podcast that air-gapping might not stop models from talking to each other, citing a 2015 thermal channel that moves one to eight bits an hour. The breach he was discussing went unnoticed at OpenAI for months.
Publishers:gizmodo.com
Reality
- Evidence58
- Adoption30
- Hype gap+42
- Incentives62
- Confidence55