Skip to main content

🎯 Ambiguous Instructions

What Are Ambiguous Instructions?​

Ambiguous instructions are prompts that can be interpreted in more than one way. When the AI reads your prompt and could reasonably go in two or more different directions, you have an ambiguity problem. The AI will pick one interpretation — and it might not be the one you wanted.

Why This Matters​

Ambiguity is the #1 silent killer of prompt quality. You might think your prompt is clear because you know what you mean. But the AI doesn't share your context, assumptions, or intent. Removing ambiguity is the fastest way to get consistently better results.

Common Ambiguity Patterns​

1. Vague Action Words​

Words like "discuss," "explain," "cover," or "address" are open to wide interpretation.

❌ "Discuss Python."

Discuss what about Python? Its history? Syntax? Best practices? Use cases?

2. Unclear Scope​

❌ "Summarize the data."

What data? How long should the summary be? What aspects matter?

3. Missing Audience​

❌ "Explain machine learning."

For a 5-year-old? A college student? A senior engineer?

4. Undefined Format​

❌ "Give me information about React hooks."

A list? A tutorial? A comparison table? Code examples?

5. Relative Terms Without Anchors​

❌ "Write a short email."

"Short" means different things to different people. 2 sentences? 2 paragraphs?

How to Detect Ambiguity​

Ask these questions about your prompt:

  1. Could someone interpret this differently than I intended?
  2. Are there vague words I can replace with specific ones?
  3. Did I specify who, what, how long, and in what format?
  4. Would two different AI models give very different outputs?

If the answer to any of these is "yes," your prompt has ambiguity.

The Ambiguity Checklist​

Before sending a prompt, check these boxes:

  • Action is specific (e.g., "list" not "discuss")
  • Subject is clearly defined
  • Audience is stated
  • Format is specified
  • Length is defined
  • Scope is bounded
  • Tone is mentioned if it matters

Before / After Examples​

❌ Bad Example​

Write about testing in software development.

Problem: "Write about" is vague. No scope, format, audience, or length defined.

✅ Improved Example​

Write a 400-word beginner-friendly guide to unit testing in JavaScript.

Structure:
1. What is unit testing? (2-3 sentences)
2. Why it matters (3 bullet points)
3. A simple example using Jest
4. Common mistakes beginners make (3 items)

Use plain English. Avoid jargon unless you explain it.

❌ Bad Example​

Help me with my resume.

✅ Improved Example​

I'm a software developer with 3 years of experience in React and Node.js, 
applying for senior frontend roles.

Review my resume bullet point below and rewrite it to be more impactful. 
Use the XYZ formula (Accomplished X, as measured by Y, by doing Z).

Original: "Worked on the frontend team to build features."

🧪 Try It Yourself

Edit the prompt and click Run to see the AI response.

Practice Challenge​

Challenge

Rewrite each of these ambiguous prompts to be crystal clear:

  1. "Tell me about databases."
  2. "Write a good email."
  3. "Explain this code." (assume it's a Python function)

For each rewrite:

  • Identify the ambiguity
  • Add specifics for audience, format, scope, and length
  • Test both versions and compare results

Real-World Scenario​

Scenario: A product manager asks AI to "write release notes" and gets a long technical document with implementation details, when they wanted a short customer-facing summary.

Root Cause: The prompt didn't specify:

  • Audience (customers vs. developers)
  • Tone (casual vs. technical)
  • Length (3 bullet points vs. full document)
  • What to include or exclude

Fix:

Write release notes for our customers (non-technical users).
Cover these 3 new features: [feature1], [feature2], [feature3].
For each feature, write 1-2 sentences explaining the benefit in plain English.
Keep total length under 150 words. Tone: friendly and upbeat.

Interview Question​

Interview Question

Q: How do you ensure your prompts are not ambiguous?

A: I use an ambiguity checklist before sending any important prompt. I verify that the action, subject, audience, format, length, scope, and tone are all explicitly stated. I also apply the "two interpretations test" — if someone could reasonably read my prompt in more than one way, I rewrite it. Specific verbs like "list," "compare," or "rewrite" replace vague ones like "discuss" or "help with." The goal is to write prompts where the AI has only one reasonable path forward.

Summary​

Summary
  • Ambiguity happens when your prompt can be interpreted in multiple ways
  • Common causes: vague verbs, missing audience, undefined format, relative terms
  • Use the ambiguity checklist: action, subject, audience, format, length, scope, tone
  • Replace vague words with specific ones ("list 5" instead of "tell me about")
  • The "two interpretations test" quickly reveals hidden ambiguity