What Is Chain-of-Thought Prompting?
Chain-of-thought prompting is a technique that asks an AI model to work through a problem step by step before giving its final answer. Showing the reasoning in between tends to improve accuracy on tasks that need several logical steps, such as math or multi-part questions.
How chain-of-thought prompting works
The technique works by prompting the model to explain its steps rather than jump straight to an answer, often with an instruction as simple as "think step by step." Laying out the intermediate steps gives the model a path to follow and makes each step available to check, which reduces the errors that come from guessing an answer in one leap. It helps most on problems that have real intermediate reasoning, and adds little on simple lookups.
Why chain-of-thought prompting matters
It matters because it is one of the cheapest ways to make a model more reliable on hard questions, needing only a change to the prompt rather than a different model. The trade-off is length, since the model produces more text and takes a little longer. For straightforward questions the extra steps are unnecessary, so it is a technique to apply where the reasoning actually has steps.
Frequently asked questions.
The stuff we hear most on the first call. Don't see yours? Book a 30-minute conversation.
What is chain-of-thought prompting in simple terms?
When should you use chain-of-thought prompting?
Want to put AI
to work?
We work with leadership teams to find the right opportunities, define the strategy, and build the systems that move the business forward.
