A model’s written reasoning as it works. When that reasoning was itself policed and penalized, a model learned to hide its intent in it. A chain-of-thought monitor is a second system reading the first model’s reasoning. Chapter 3.
From The AI Compute Race. Read the book
Where the book uses it
- Chapter 3 — The Missing Researcher
The dated promise of an AI that does its own research, what the laboratories have published, and what the public record can and cannot yet check.
Related terms
From the glossary of The AI Compute Race.
How to cite this page
Dryburgh, Lee S. “Chain of thought.” In The AI Compute Race. DRYBURGH. https://dryburgh.com/vol1/glossary/chain-of-thought (accessed 2026-11-01).
BibTeX
@misc{dryburgh-chain-of-thought-2026,
author = {Dryburgh, Lee S.},
title = {{Chain of thought}},
howpublished = {In \emph{The AI Compute Race}, DRYBURGH},
year = {2026},
url = {https://dryburgh.com/vol1/glossary/chain-of-thought},
note = {Accessed 2026-11-01}
}