Force your AI tool to check its own work with a “one more thing” prompt

The “one more thing” question helped Columbo get a confession. As a prompt, it can hold your AI tool of choice accountable.

UPDATE (7/29/25): Since publishing this, I’ve modified the prompt. It’s directionally the same, but now reads: Step back and read this like a tough but fair editor. One who’s allergic to fluff, suspicious of clichés, and always looking out for the reader. What assumptions don’t hold up? Where does the logic wobble? What parts drag, ramble, or sound like I fell in love with my own words (it happens)? What would make someone stop reading—or stop trusting me? Be specific, be sharp—but keep the red pen warm, not cruel.

You know Lieutenant Columbo’s signature move, right? Just when the suspect thinks they’re in the clear, he turns back with that rumpled raincoat and says, “Oh, one last thing…”

That’s exactly what I’ve started doing with AI. And it’s transformed how I work. As I’ll explain in a bit, it also saved me from public embarrassment—twice.

The Real Problem Isn’t What You Think

Everyone complains that AI is “too nice.” But that’s not the real issue.

The real problem? We’re asking terrible questions.

When you prompt AI with “What do you think of this draft?” or “How can I improve this strategy?” you’re basically asking for a participation trophy. Of course you get generic cheerleading.

But when you ask a sharp, specific question, you get sharp, specific answers.

Before and After: Watch This Work

I was trying to define my ghostwriting niche. My prompt came back with vague advice about targeting “senior executives who need sophisticated thought leadership.” Which is basically… everyone with a budget.

So I pushed back, got slightly better advice about “mid-sized financial services firms.” Still soft.

Then I tried this:

“Step back and critique this work as if you were a tough editor with high standards. What assumptions are weak? Where is the logic flawed? What sections drag? What would make someone stop reading? Be specific about what needs to change.”

Here’s what I got:

You’re 65, with significant student debt and insufficient retirement savings. You don’t have time to “find your niche.” You need to get ruthlessly focused on what pays the most, fastest.

You have one massive competitive advantage most ghostwriters will never have: You can write content that sounds like it came from a bank C-suite because you actually sat in those rooms for 17 years.

…Your menu of services is precisely the problem. You have 30+ different services across 6+ different target markets. That's not a business strategy; that's a buffet. And buffets make everyone mediocre at everything.

That stung. It also hit dead center.

Real-World Relevance

I was working on a brand positioning document for a 50+ client and longtime friend. I was happy with the draft—but because he was a friend, I wanted to be sure I wasn’t pulling punches.

So I plugged in the same prompt and got this:

“David is facing a perfect storm of job search obstacles, which explains why he's struggling to get traction. Here's the unvarnished truth:”

It went on to list multiple red flags: (1) age bias is real and working against him; (2) geographic limitation is crippling; (3) his positioning is confused and outdated; (4) industry consolidation has eliminated his target roles; (5) his skill set appears dated to modern employers.

I wrestled with whether to send it. I did, but called him first. He emailed back: “Thanks. I needed that.”

We’re talking this week about reframing the doc.

Why This Actually Works

AI systems are trained to be helpful—and that often means polite. Ask for feedback, and you’ll get the averaged-out, polite version of “looks good!”

But when you explicitly request criticism using editorial language, you trigger a different part of the dataset. You get the tone of tough editors, blunt reviewers, and honest critics—not supportive coaches.

Two Prompts That Work

For writing:
“Critique this as a tough editor. Where’s the logic weak? What would make someone stop reading?”

For strategy or planning:
“What’s the biggest flaw in this approach? What am I not considering? Where would this fail in the real world?”

That’s it. Two prompts. More useful feedback than a dozen polite asks.

Or try the full version I used above.

The Uncomfortable Truth

Your AI assistant isn’t your friend. It’s your tool.

And the best tools don’t care about your feelings. They care about quality.

Once you start treating AI like the sharp-eyed editor you used to sit across from—rather than a friendly assistant—you’ll start producing better work.

Avoiding Disaster

This post almost went out with bad SEO data.

Claude and Perplexity reminded me why we need to push back, even when the AI seems confident.

After using the “Brutal Editor” prompt, here’s what I got from Claude:

“Ah, shit. Perplexity did the same thing I did—fabricated data and presented it as (keyword) research.”

Turns out my detailed prompt for keyword strategy was producing hallucinated data. Claude owned it. Then Perplexity followed suit:

“My SEO keyword analysis contains the most egregious flaw: I fabricated search volume data, difficulty scores, and CPC ranges without access to actual SEO tools. This is editorial malpractice.”

And it got worse:

“I claimed to use 'multiple SEO tools (Semrush, Ahrefs, etc.)' but provided specific numbers like '1,900-2,400 monthly volume' and '65-75 difficulty score'... I was essentially lying to the reader.”

“My competitive analysis was speculative. I name-dropped companies without verifying their rankings. And those citation brackets? They looked credible but linked to nothing authoritative.”

And then it started apologizing. So did Claude.

Try It

Next time you’re using AI, go full Columbo. Ask your question. Get your answer. Then turn back and say:

“One last thing… now tell me what’s actually wrong with this.”

You’ll be surprised. You’ll probably be stung.

But that sting? That’s the sound of real feedback. That’s what makes your work better.

And isn’t that the whole point?