Zac Kenton
@ZacKenton1
Amplified Oversight team lead @GoogleDeepMind | AGI safety & alignment | Enabling accurate human supervision of superhuman AI.
Joined May 2014
1.6K Following 2.3K Followers
Great work on using probes for classifying real-world misuse risks!
Our new @GoogleDeepMind paper studies novel activation probe architectures for classifying real-world misuse risks.
Our research has informed live deployments of probes in Gemini. 🧵
Show more
