Chain-of-Thought Reasoning
When an AI model works through a problem in visible, step-by-step reasoning before giving a final answer.
01 Definition
Chain-of-thought reasoning is a technique where an AI model generates intermediate reasoning steps — 'thinking out loud' — before producing its final answer, which tends to improve accuracy on math, logic, and multi-step problems compared to answering immediately.
02 Why It Matters
This is why some AI models show a visible 'thinking' process, and why asking a model to 'think step by step' often improves the quality of its answers on hard problems.
03 Example
Asking 'solve this step by step' instead of just 'what's the answer' often produces a more accurate result on math word problems.