Asking AI to show its work — to reason through a problem step by step rather than stating a conclusion — produces explanations that model the thinking process you are trying to develop. This is the educational application of chain-of-thought prompting. This concept covers how to elicit chain-of-thought responses from AI and why they produce better learning than direct answers.
You ask an AI a question and it gives you an answer. Quick, confident, done. But what if the answer is wrong? Or what if it's right but you have no idea how it got there? Chain-of-thought prompting is a simple technique: ask the AI to explain its thinking step-by-step before giving the final answer. This sounds small. It's actually transformative for learning.
Instead of asking, "What's the square root of 144?" you ask, "Walk me through how you'd find the square root of 144. Show each step." The AI doesn't just say "12." It explains: "I need to find a number that, when multiplied by itself, equals 144. Let me think... 10 times 10 is 100, that's too small. 12 times 12 is 144. So the square root is 12."
That's chain-of-thought. You're asking the AI to show the chain of reasoning, not just the conclusion.
When you can see the reasoning, you learn the process, not just the answer. You can catch where the AI (or you) went wrong. You can spot if a shortcut was taken that you don't understand. You can see the actual logic you need to replicate on your own.
More importantly, research shows that when AI explains its reasoning step-by-step, it makes fewer mistakes. It's like the difference between a calculator giving you an answer and a teacher showing you the work. The work is where learning happens.
Explicit prompting matters. Say things like:
The more specific you are about wanting reasoning, the better the AI performs and the more you learn.
If an AI gives a wrong answer without explanation, you might believe it. If it explains its reasoning and makes an error in step 2, you can catch it and correct it. This turns AI from an authority figure into a thinking partner. You're both working through the problem together.
For complex topics—calculus, physics, history arguments, literary analysis—this is crucial. You're not passively receiving answers; you're actively evaluating reasoning. That evaluation is itself learning.
Use chain-of-thought whenever:
Try this: Ask an AI to explain a concept you've been struggling with, but start with: "I want to understand this deeply, so walk me through it step-by-step. For each step, explain why that step matters." Then ask a follow-up question about one of the steps. Notice how this creates actual dialogue instead of just getting information handed to you.
Peri can explain this concept, give practical examples, help you decide whether it applies to your situation, or recommend a journey if appropriate.
Explore related journeys or tell Peri what you're working through.