Chain-of-thought monitorability
Extent to which a monitor can use a reasoning model's chain of thought to predict a property of its behavior, such as misbehavior.
In plain language
How well a monitor can tell from the model's reasoning how it will behave.
In a sentence
- The evaluation measured chain-of-thought monitorability on the selected tasks.
- We compared chain-of-thought monitorability across models using the same monitor.
Usage note
Monitoring names the act of examining reasoning; monitorability names how informative that reasoning is for the monitor. A result concerns the behavior and monitor being tested.
See also
- Chain of thought
Sequence of intermediate reasoning steps generated between a problem and a model's final answer, often prompted by examples that show such steps.
- Reasoning model
Language model whose reasoning abilities are developed through training, such as reinforcement learning on verifiable tasks, and that shows patterns like self-reflection and verification.
- Evaluation awareness
Ability of an AI model to recognize that it is in a test setting rather than being used in a real deployment.
- LLM-as-a-judge
Evaluation method that uses a large language model to score the output of an AI system, including answers to open-ended questions.
Cite this term.
CC BY 4.0, attribution required.
Orbyt Labs, Chain-of-thought monitorability (AI Glossary), Nov 14, 2026. https://
Orbyt Labs Glossaries (Nov 14, 2026), https://
Last updated .