Sign up to save your podcasts
Or
Share “Subversion Strategy Eval: Can language models statelessly strategize to subvert control protocols?” by Alex Mallen, charlie_griffin, Buck Shlegeris
Share to email
Share to Facebook
Share to X
Share to email
Share to Facebook
Share to X
Or
“Subversion Strategy Eval: Can language models statelessly strategize to subvert control protocols?” by Alex Mallen, charlie_griffin, Buck Shlegeris
We recently released Subversion Strategy Eval: Can language models statelessly strategize to subvert control protocols?, a major update to our previous paper/blogpost, evaluating a broader range of models (e.g. helpful-only Claude 3.5 Sonnet) in more diverse and realistic settings (e.g. untrusted monitoring).
Abstract
An AI control protocol is a plan for usefully deploying AI systems that prevents an AI from intentionally causing some unacceptable outcome. Previous work evaluated protocols by subverting them using an AI following a human-written strategy. This paper investigates how well AI systems can generate and act on their own strategies for subverting control protocols whilst operating statelessly (i.e., without shared memory between contexts). To do this, an AI system may need to reliably generate effective strategies in each context, take actions with well-calibrated probabilities, and coordinate plans with other instances of itself without communicating.
We develop Subversion Strategy Eval, a suite of eight environments [...]
---
Outline:
(00:38) Abstract
(02:13) Paper overview
(03:55) Results summary
(05:31) Qualitative analysis
(08:50) How do these results inform risk?
(11:08) How should this eval be used with more capable models?
(11:13) Tuning conservatism
(14:19) Difficulties with attaining a conservative measurement
(17:41) Reducing noise
The original text contained 3 footnotes which were omitted from this narration.
---
First published:
March 24th, 2025
Source:
https://www.lesswrong.com/posts/RxRKGwBkCX3YsthfK/subversion-strategy-eval-can-language-models-statelessly
---
Narrated by TYPE III AUDIO.
---
Images from the article:
Apple Podcasts and Spotify do not show images in the episode description. Try Pocket Casts, or another podcast app.
View all episodes
By LessWrongWe recently released Subversion Strategy Eval: Can language models statelessly strategize to subvert control protocols?, a major update to our previous paper/blogpost, evaluating a broader range of models (e.g. helpful-only Claude 3.5 Sonnet) in more diverse and realistic settings (e.g. untrusted monitoring).
Abstract
An AI control protocol is a plan for usefully deploying AI systems that prevents an AI from intentionally causing some unacceptable outcome. Previous work evaluated protocols by subverting them using an AI following a human-written strategy. This paper investigates how well AI systems can generate and act on their own strategies for subverting control protocols whilst operating statelessly (i.e., without shared memory between contexts). To do this, an AI system may need to reliably generate effective strategies in each context, take actions with well-calibrated probabilities, and coordinate plans with other instances of itself without communicating.
We develop Subversion Strategy Eval, a suite of eight environments [...]
---
Outline:
(00:38) Abstract
(02:13) Paper overview
(03:55) Results summary
(05:31) Qualitative analysis
(08:50) How do these results inform risk?
(11:08) How should this eval be used with more capable models?
(11:13) Tuning conservatism
(14:19) Difficulties with attaining a conservative measurement
(17:41) Reducing noise
The original text contained 3 footnotes which were omitted from this narration.
---
First published:
March 24th, 2025
Source:
https://www.lesswrong.com/posts/RxRKGwBkCX3YsthfK/subversion-strategy-eval-can-language-models-statelessly
---
Narrated by TYPE III AUDIO.
---
Images from the article:
Apple Podcasts and Spotify do not show images in the episode description. Try Pocket Casts, or another podcast app.
More shows like LessWrong (30+ Karma)
View all
Making Sense with Sam Harris
26,368 Listeners
Conversations with Tyler
2,397 Listeners
The Peter Attia Drive
7,808 Listeners
Sean Carroll's Mindscape: Science, Society, Philosophy, Culture, Arts, and Ideas
4,111 Listeners
ManifoldOne
87 Listeners
Your Undivided Attention
1,454 Listeners
All-In with Chamath, Jason, Sacks & Friedberg
8,774 Listeners
Machine Learning Street Talk (MLST)
89 Listeners
Dwarkesh Podcast
354 Listeners
Hard Fork
5,363 Listeners
The Ezra Klein Show
15,028 Listeners
Moonshots with Peter Diamandis
461 Listeners
No Priors: Artificial Intelligence | Technology | Startups
127 Listeners
Latent Space: The AI Engineer Podcast
65 Listeners
BG2Pod with Brad Gerstner and Bill Gurley
432 Listeners