Our latest paper shows that unsupervised methods on LLM activations don’t yet discover latent knowledge. Many things can satisfy knowledge-like properties besides ground truth. E.g a strongly opinionated character causes ~half the probes to detect *her* beliefs instead.

I had fun talking to Daniel on his podcast AXRP! And I’ve enjoyed listening to his other episodes too :)

Very cool find by Senthooran Rajamanoharan, Arthur Conmy, and the rest of the DeepMind mechinterp team! I’m excited by the rate of progress here.

Excited to see what people try with these shiny new open source SAEs! Great work by Senthooran Rajamanoharan and the team on pushing SOTA here

New Google DeepMind safety paper! LLM agents are coming – how do we stop them finding complex plans to hack the reward? Our method, MONA, prevents many such hacks, *even if* humans are unable to detect them! Inspired by myopic optimization but better performance – details in🧵

RL training can incentivise LLM agents to produce long-term alien plans, and evade monitoring. But in high-stakes settings, comprehensibility is critical. Our new paper shows how to change an agent’s incentives to *only* act in ways that we can understand.

We are excited to release a short course on AGI safety! The course offers a concise and accessible introduction to AI alignment problems and our technical & governance approaches, consisting of short recorded talks and exercises (75 minutes total). deepmindsafetyresearch.medium.com/1072adb7912c

We're hiring! Join an elite team that sets an AGI safety approach for all of Google -- both through development and implementation of the Frontier Safety Framework (FSF), and through research that enables a future stronger FSF.