Productivity Tip: Fact Check Prompt
Let’s talk about something we’ve all faced: You ask your AI (Copilot, ChatGPT, Gemini, whatever) a question, and it gives you a beautifully worded, confident answer. But… can you trust it?
The reality is that while AI can sound authoritative, it’s only as good as the data behind it. And that’s where the Fact Check prompt pattern comes in. It’s a simple, repeatable method to help you validate the AI’s answers—and boost the quality of your work in the process.
Step 1: Start with an Outline
In this example, I’m starting with a simple productivity task: asking AI to generate (or expand upon) an outline on a topic I care about. Why an outline? So it’s easier to demonstrate this pattern and validate the results—but feel free to try it with any content. This is a great warm-up prompt. You’re not committing to final content yet—just exploring ideas.
The outline was helpful and well-organized. But here’s where most people stop—and where mistakes can begin. If I used this content as-is, I’d have no idea whether it’s grounded in fact or just well-crafted filler.
Step 2: Go Deeper on Specific Points
To see how well AI could support each part of its outline, I asked it to expand on individual sections. This gave me more detail—but also raised more questions. How do I know these claims are accurate? Are they based on real data or just assumptions?
This is where the Fact Check pattern starts to shine.
Step 3: Ask Copilot to List Its Facts
Here’s the magic prompt:
While my example is a little wordy, this set of instructions turns a passive interaction into an active validation loop. It forces the AI to surface what it’s really working from—whether that’s real data, common knowledge, or just inferred logic.
Now, instead of guessing, I can critically assess:
- Are these facts correct?
- Are they being used appropriately in the AI’s conclusions?
Step 4: Keep Pushing for Granularity
If the facts are too vague or generalized, ask AI to go deeper. In my case, I followed up by requesting more granular data, and to cite specific resources used. The quality improved, and so did my confidence in the results.
Each round of this pattern gives you a more transparent view—and often better ideas too.
Why This Pattern Works
The Fact Check pattern isn’t just a “gotcha” tactic for catching AI errors. It’s a productivity booster. By developing the habit of asking why an answer is what it is, you:
- Avoid costly rework later
- Learn faster (and often spot holes in your own understanding)
- Model better habits for your team when working with AI
It’s a small step with big returns. So the next time you use AI, build in this habit:
- Get the answer
- Ask for the facts behind it
- Challenge it where needed
- Repeat until you trust it
Because when it comes to productivity, truth is a feature—and you’re in charge of enabling it.
