Critically review an AI answer before I rely on it
Stress-tests an AI answer for hidden assumptions, missing context, weak reasoning, and the strongest counterargument.
The prompt
Critically evaluate an AI-generated answer before I rely on it. Be a tough, fair reviewer.
The question I asked: [PASTE the original prompt or question]
The AI answer: [PASTE]
What I'll use it for: [DECISION or DELIVERABLE, and how much is at stake]
Context the AI may not have had: [e.g. our budget, history, team, market, constraints]
Review it for:
1. Fit - does it actually answer my question, or a slightly different, easier one?
2. Assumptions - list every assumption it makes, stated or implied, and rate each as reasonable / questionable / wrong for my situation.
3. Missing context - what it ignores or could not know that would change the answer.
4. Weak reasoning - leaps in logic, vague generalities, one-sided framing, or advice that sounds good but doesn't hold up.
5. Strongest counterargument - the best case against the answer's main conclusion, argued properly.
6. What would change the conclusion - the facts or conditions that would make this the wrong answer.
Return:
1. A verdict in one line: rely on it / use with changes / don't use.
2. A table: Issue | Where (short quote) | Type (assumption / missing context / weakness) | Severity | Fix.
3. The strongest counterargument in one paragraph.
4. An improved version of the answer's key conclusion, or the follow-up questions I should ask.
Don't nitpick wording; focus on what changes the outcome. Flag factual claims that need checking, but don't present your own corrections as certain without a source.
Highlighted [brackets] are blanks to replace with your own details.
Fill in the blanks and use this promptMore Working with AI prompts
What to fill in
- PASTE the original prompt or question
- PASTE
- DECISION or DELIVERABLE, and how much is at stake
- e.g. our budget, history, team, market, constraints
How this prompt is built
This prompt follows the RTF structure (Role, Task, Format). RTF tells the model who to be, what to do and what shape the answer should take. It is the quickest structure that still gives noticeably better results than a bare request.
Related prompts
- Reverse-engineer the prompt behind an outputWorks backwards from text, code, or an image to the prompt that likely produced it.
- List the ways an AI agent could fail and how to prevent eachA failure-mode review for an AI agent: realistic scenarios, how you would notice, the damage, and the safeguard.
- Answer my question, then critique it and argue against itGets an answer, an honest self-critique, the strongest counterargument, and a revised answer you can trust more.
- Turn a one-off AI chat into a reusable prompt pack for my teamExtracts the prompts, fixes, and checks from an AI chat that worked into a step-by-step workflow and prompt pack colleagues can reuse.