See the Difference

Same question. One gives permission to be uncertain.

Without permission
You: "What year did Johnson & Murphy start making shoes in Nashville?"

Johnson & Murphy was founded in 1850 and began manufacturing in Nashville in 1882, where they established their primary production facility.

Sounds confident. But is it true? AI gave specific details whether or not it actually knows.

With permission
You: "What year did Johnson & Murphy start making shoes in Nashville? If you're not certain, say so."

I'm not certain about the exact year they started manufacturing in Nashville. Johnson & Murphy was founded in 1850, but I'd recommend checking their official history for specific details about the Nashville timeline.

Honest. Now you know what to trust and what to verify.

Why This Works

AI is trained to be helpful, which means it almost always gives an answer—even when it shouldn't. If it doesn't know something, it guesses. And because AI sounds confident no matter what, those guesses look exactly like facts.

When you give AI permission to say "I don't know," you change the dynamic. Instead of making something up, it can admit uncertainty.

Ways to Use This

The Technique

Add "if you're not sure, say so" to any prompt where accuracy matters. AI will stop guessing and start being honest about what it actually knows.

When to Use This