Chain-of-Thought: Make the AI Show Its Work

Telling the AI to 'think step by step' before answering dramatically improves its accuracy on reasoning problems. Here's why and when.

24 min · Reviewed 2026

The four magic words

Adding 'think step by step' or 'let's work through this carefully' to a prompt can flip an AI from wrong to right on math, logic, and planning problems. Researchers call this technique chain-of-thought, or CoT.

With and without CoT

WITHOUT CoT: Q: A shirt costs $20 and is 25% off. Then you use a $5 coupon on the discounted price. What do you pay? A: $10. <-- often wrong WITH CoT: Q: A shirt costs $20 and is 25% off. Then you use a $5 coupon on the discounted price. What do you pay? Think step by step. A: - Original price: $20. - 25% of $20 is $5. - Discounted price: $20 - $5 = $15. - Apply $5 coupon: $15 - $5 = $10. Final answer: $10. <-- shows reasoning, more reliably correctCoT turns a guess into a verifiable chain.

Here the final number is the same, but with CoT you can check each step. If a step is wrong, you see it. Without CoT, the AI could give you '$15' and you wouldn't know why.

Phrasings that trigger CoT

'Think step by step.'
'Let's reason through this carefully.'
'Before giving your final answer, write out your reasoning.'
'Work it out one piece at a time.'
'Show your thinking as a numbered list, then give the answer.'

When to use CoT

Use CoT	Skip CoT
Multi-step math.	Quick factual lookup.
Logic puzzles.	Creative writing.
Planning and scheduling.	Short answers or summaries.
Comparing options.	Simple translations.
Anywhere getting it wrong would matter.	Anywhere speed matters more than accuracy.

You are a careful tutor. A student asks: "I have 3 tests coming up. Math is worth 20%, Science is worth 35%, and History is worth 45% of my grade. If I scored 92, 78, and 85 respectively, what's my final grade?" Before you give a final number: 1. Write out what each score contributes. 2. Show the math. 3. Only then give the total. Format: **Step 1:** **Step 2:** **Step 3:** **Final grade:**Structured CoT — you can verify each step.

End-of-lesson check

8 questions · take it digitally for instant feedback at tendril.neural-forge.io/learn/quiz/end-prompting-chain-of-thought-builders

What is the main idea of "Chain-of-Thought: Make the AI Show Its Work"?
1. Telling the AI to 'think step by step' before answering dramatically improves its accuracy on reasoning problems. Here's why and when.
2. Use AI as the final authority for the whole decision
3. Avoid checking the answer once it sounds polished
4. Focus only on speed instead of judgment
Which concept is most central to "Chain-of-Thought: Make the AI Show Its Work"?
1. step-by-step reasoning
2. chain-of-thought
3. show your work
4. reasoning
Which use of AI fits this topic best?
1. Let the AI decide what matters without your review
2. Use the answer before checking whether it fits the situation
3. 'Think step by step.'
4. Use the first answer without checking it
What should a careful learner remember about "Why it works"?
1. Use AI to draft or organize ideas about chain-of-thought, then verify before acting.
2. Skip the context so the tool can guess faster
3. Treat the output as private even after sharing it online
4. Use the answer without checking the source
You want to use AI after this lesson. What is the safest next step?
1. Act immediately because the AI answer is written clearly
2. Use the AI answer as a draft, then check it against a reliable source.
3. Hide uncertainty so the final answer looks cleaner
4. Use private or sensitive details before checking permission
How should AI output about chain-of-thought be treated?
1. As proof that no other source is needed
2. As a replacement for context, consent, or expert review
3. As a draft or helper output that still needs human judgment and verification
4. As something that becomes correct when it sounds confident
Name one way to verify an AI answer about chain-of-thought.
Which action would help you apply "Chain-of-Thought: Make the AI Show Its Work" responsibly?
1. Use the tool to avoid thinking through the tradeoff
2. Keep going even if the output conflicts with a trusted source
3. Use the first answer without checking it
4. 'Let's reason through this carefully.'

← Back to interactive lesson

Tendril · Builders · Prompting

Chain-of-Thought: Make the AI Show Its Work

Telling the AI to 'think step by step' before answering dramatically improves its accuracy on reasoning problems. Here's why and when.

24 min · Reviewed 2026

The four magic words

With and without CoT

WITHOUT CoT: Q: A shirt costs $20 and is 25% off. Then you use a $5 coupon on the discounted price. What do you pay? A: $10. <-- often wrong WITH CoT: Q: A shirt costs $20 and is 25% off. Then you use a $5 coupon on the discounted price. What do you pay? Think step by step. A: - Original price: $20. - 25% of $20 is $5. - Discounted price: $20 - $5 = $15. - Apply $5 coupon: $15 - $5 = $10. Final answer: $10. <-- shows reasoning, more reliably correctCoT turns a guess into a verifiable chain.

Here the final number is the same, but with CoT you can check each step. If a step is wrong, you see it. Without CoT, the AI could give you '$15' and you wouldn't know why.

Phrasings that trigger CoT

'Think step by step.'
'Let's reason through this carefully.'
'Before giving your final answer, write out your reasoning.'
'Work it out one piece at a time.'
'Show your thinking as a numbered list, then give the answer.'

When to use CoT

Use CoT	Skip CoT
Multi-step math.	Quick factual lookup.
Logic puzzles.	Creative writing.
Planning and scheduling.	Short answers or summaries.
Comparing options.	Simple translations.
Anywhere getting it wrong would matter.	Anywhere speed matters more than accuracy.

You are a careful tutor. A student asks: "I have 3 tests coming up. Math is worth 20%, Science is worth 35%, and History is worth 45% of my grade. If I scored 92, 78, and 85 respectively, what's my final grade?" Before you give a final number: 1. Write out what each score contributes. 2. Show the math. 3. Only then give the total. Format: **Step 1:** **Step 2:** **Step 3:** **Final grade:**Structured CoT — you can verify each step.

End-of-lesson check

8 questions · take it digitally for instant feedback at tendril.neural-forge.io/learn/quiz/end-prompting-chain-of-thought-builders

What is the main idea of "Chain-of-Thought: Make the AI Show Its Work"?
1. Telling the AI to 'think step by step' before answering dramatically improves its accuracy on reasoning problems. Here's why and when.
2. Use AI as the final authority for the whole decision
3. Avoid checking the answer once it sounds polished
4. Focus only on speed instead of judgment
Which concept is most central to "Chain-of-Thought: Make the AI Show Its Work"?
1. step-by-step reasoning
2. chain-of-thought
3. show your work
4. reasoning
Which use of AI fits this topic best?
1. Let the AI decide what matters without your review
2. Use the answer before checking whether it fits the situation
3. 'Think step by step.'
4. Use the first answer without checking it
What should a careful learner remember about "Why it works"?
1. Use AI to draft or organize ideas about chain-of-thought, then verify before acting.
2. Skip the context so the tool can guess faster
3. Treat the output as private even after sharing it online
4. Use the answer without checking the source
You want to use AI after this lesson. What is the safest next step?
1. Act immediately because the AI answer is written clearly
2. Use the AI answer as a draft, then check it against a reliable source.
3. Hide uncertainty so the final answer looks cleaner
4. Use private or sensitive details before checking permission
How should AI output about chain-of-thought be treated?
1. As proof that no other source is needed
2. As a replacement for context, consent, or expert review
3. As a draft or helper output that still needs human judgment and verification
4. As something that becomes correct when it sounds confident
Name one way to verify an AI answer about chain-of-thought.
Which action would help you apply "Chain-of-Thought: Make the AI Show Its Work" responsibly?
1. Use the tool to avoid thinking through the tradeoff
2. Keep going even if the output conflicts with a trusted source
3. Use the first answer without checking it
4. 'Let's reason through this carefully.'

← Back to interactive lesson