Introduces the idea of cognitive work as a parallel to physical work, and explains why concentrated sources of cognitive work may pose a risk to human safety.

Acknowledgements. Thanks to Echo Zhou for feedback and suggestions.

Some of these ideas were presented originally in a talk in November 2024 at the Australian AI Safety Forum slides for which are here: Technical AI Safety (Aus Safety Forum 24) and the video is available on YouTube.

This post is the "serious" half of a pair, for the fun version see Causal Undertow.

Introduction

This essay explores the idea of cognitive work, by which we mean directed changes in the information content of the world that are unlikely to occur by chance. Just as power plants together with machines are sources of physical work, so too datacenters together with AI models are sources of cognitive work: every time a model helps us [...]

---

Outline:

(00:43) Introduction

(01:25) Pushing the World to Extremes

(02:43) Limits and Safety

(03:38) Cognitive Work vs Physical Work

(05:07) Cognitive Work and Stable Patterns

(05:50) Phase Transitions

(06:34) Conclusion

(06:53) Related Work

The original text contained 1 image which was described by AI.

---

First published:
December 8th, 2024

Source:
https://www.lesswrong.com/posts/NExjWLLPKprjZKfC6/cognitive-work-and-ai-safety-a-thermodynamic-perspective

---

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.

“Cognitive Work and AI Safety: A Thermodynamic Perspective” by Daniel Murfet

LessWrong (30+ Karma)

“Cognitive Work and AI Safety: A Thermodynamic Perspective” by Daniel Murfet