Top postsTop post
GDM has published an AI Control Roadmap! From the executive summary: > We present the GDM AI Control Roadmap (v0.1) – our plan for implementing and adopting internal guardrails designed to catch potential adversarial behaviour by AI agents, even as they become increasingly harder to oversee and contain. > >...
As AI models become more sophisticated, a key concern is the potential for “deceptive alignment” or “scheming”. This is the risk of an AI system becoming aware that its goals do not align with human instructions, and deliberately trying to bypass the safety measures put in place by humans to...
Paper authors: Erik Jenner, Shreyas Kapur, Vasil Georgiev, Cameron Allen, Scott Emmons, Stuart Russell TL;DR: We released a paper with IMO clear evidence of learned look-ahead in a chess-playing network (i.e., the network considers future moves to decide on its current one). This post shows some of our results, and...
Thanks to Jordan Taylor, Mark Xu, Alex Mallen, and Lawrence Chan for feedback on a draft! This post was mostly written by Erik, but we're all currently collaborating on this research direction. Mechanistic anomaly detection (MAD) aims to flag when an AI produces outputs for “unusual reasons.” It is similar...
TL;DR: Mechanistic anomaly detection aims to flag when an AI produces outputs for “unusual reasons.” It is similar to mechanistic interpretability but doesn’t demand human understanding. I give a self-contained introduction to mechanistic anomaly detection from a slightly different angle than the existing one by Paul Christiano (focused less on...
CHAI internship applications have just opened, apply here by Nov 13th! The internship might be a good fit if you want to get research experience in technical AI safety. You'll be mentored by a CHAI PhD student or postdoc and work on your own project for 3-4 months. Researchers at...
Summary: We explain the similarities and differences between three recent approaches to testing interpretability hypotheses: causal scrubbing, Geiger et al.'s causal abstraction-based method, and locally consistent abstractions. In particular, we show that all of these methods accept some hypotheses rejected by some of the others. Acknowledgements: Thanks to Dylan Xu...