What is Chain-of-thought prompting?

Chain-of-thought prompting asks an AI model to reason step by step before giving its answer, which improves accuracy on problems that need working-out.

By adding something as simple as "think through this step by step," you nudge the model to lay out its reasoning before committing to a conclusion. For multi-step problems — maths, logic, planning, anything where a leap to the answer goes wrong — that explicit working noticeably raises the hit rate, and it makes the answer easier to check.

It's most useful when a task has intermediate steps that matter; for simple lookups or rewrites it just adds length. Many newer reasoning models now do this internally, so you may not need to ask — but for everyday models on a hard question, prompting for the steps is still a reliable quality lever.