Map of Verification
Question the output
When the output is "too good", you need to check quality, consistency, bias. You're about to use a result critically.
Key Question
Why might I NOT trust this answer?
Purpose
When: the output is "too good", you need to check quality, consistency, bias, you're about to use a result critically.
What It Enables
Critical prompts, Fact-check prompts, Counter-argument prompts
Risks
Passive acceptance, False competence, Silent errors
When Not to Use
In early creative phases. When you're just exploring.
Related Prompts
Bias Prompt
Identify cognitive biases in the output
Learning Prompt
Learn from the exchange to improve future prompts
Editorial fact-checking prompt
Verify content before publication by identifying fragile claims.
Practical Verification Prompt
Identify the first signal of error in application
Critical Prompt
Assume the answer is wrong and find where it fails
"When to stop" prompt
Recognize when continuing to prompt no longer improves the result.
Counter-Argument Prompt
Build the best possible objection
Reliability Prompt
Assess how reliable the answer is
Strategic code review prompt
Review code not just for bugs, but for architectural decisions and tech debt.
Output evaluation prompt
Create an evaluation rubric to judge AI output quality.