How to Use AI to Find Your Blind Spots (The Prompt That Shows You What You Cannot See)

You already know there are things about you that everyone else can see and you cannot. Here is how to use AI to find your blind spots: give it real context about who you are first, then ask it to name the hidden and non-obvious things you may be hiding from yourself. Ask a blank chat window and you get a horoscope. Ask a model that holds your goals, your history, your business, and your own words back to you, and it will name the thing you have been walking past for years.

A founder I coach did exactly this on a walk last week. He had been stuck on an assignment I gave him and he was frustrated about it. So he talked it through with AI while he walked, then took what came back and asked it to go deeper. What came back was a list. Fear. Distraction. Old stories. Limiting beliefs. Excuses. And one word set in capital letters that the others were not. COMFORT. He read that to me on our call. Twenty minutes later, talking about a goal he had been chasing for a year, he said the words "I am very comfortable stating that." He did not hear it. AI had already told him.

What Does It Mean to Use AI to Find Your Blind Spots?

A blind spot is not a fact you are missing. It is a pattern you are too close to see. You are inside the sentence, so the sentence sounds like a description of reality instead of a choice you keep making.

When you use AI to find your blind spots, you are using it the way a good coach works. A coach does not hand you new information. A coach plays your own words back to you at a volume you cannot ignore. AI can do a version of that at any hour, for as long as you want, without getting tired of you.

The founder above is a good example. He came into our call apologizing for not building the routines I had asked him to build. He had walked through it with AI first, and AI had told him something he did not expect. Every element of the operations dashboard he had already built updates on a cycle. Daily. Weekly. Monthly. Quarterly. Annually. Those are routines. He had built the whole thing months ago and never called it what it was.

His words on the call: "I always just had it framed differently than how I thought it was."

He was not behind. He was mislabeled. That is a blind spot, and it cost him weeks of feeling like he was failing at something he had already done.

Why Can You Not See Your Own Blind Spots?

Because the blind spot is doing a job for you. It is protecting something.

Most of the time that something is comfort. It is comfortable to describe where you are as a permanent fact. It costs nothing. It asks nothing of you tomorrow.

On that same call, I heard the founder say a number about himself. He did not say "I weigh that number." He said "I am that number." Almost everyone says it that way and nobody hears it. So I stopped and asked him whether he was carrying that number as his identity rather than as a measurement.

He sat with it and said yes.

Then he said the sentence that ended the argument. "That number is not where I want to go, but I am very comfortable stating it."

That is the blind spot. Not the number. The comfort with saying the number. He had already closed most of the distance to his goal, and he was measuring from the day he started rather than from where he was standing, which made the remaining gap feel bigger than the one he had already closed. It was smaller. He could not see it because he was inside it.

You run the same math on your business. Ask yourself where you keep describing a constraint as a fact.

What Is the Best Prompt to Use AI to Find Your Blind Spots?

Here is the one that produced everything above.

Then follow it with one more turn.

The second turn is where the value is. The first answer is usually the polite version. The second is the one that lands.

A framing question also works well as an opener, because it changes what the model is looking for. Try this one before you ask anything else.

What came back for this founder was worth the whole exercise. Fear, distraction, comfort, and old stories do not create who you are. They cover who you are. Systems, dashboards, and frameworks are multipliers on top of the person. The person is the source. If every system he built disappeared tomorrow, he would build them again, because the builder is what is actually there.

What Context Does AI Need to Give You an Honest Answer About Yourself?

This is the part people skip, and it is the reason most attempts at this produce nothing.

"Knowing what you know about me" is a load-bearing phrase. If the AI knows nothing about you, it will invent something generic and flattering. The quality of the answer is capped by the quality of what you have given it.

So before you run the prompt, give it the raw material:

• Your goals for the year, in your own words, including the ones you have not hit
• Your actual history, including the things that did not work
• Your values and how you make decisions
• Transcripts or notes from real conversations where you were thinking out loud
• The numbers you track, including the ones you avoid looking at

This is the same reason a second brain it can actually read beats a clever prompt every time. Context compounds. Once it is written down and stored somewhere the AI can reach, every future conversation starts further along, and you stop starting from scratch every time you open a window.

What you put in is what you get to co-create with. Put in gold, get gold back.

How Do You Audit Your Relationship With Comfort?

Here is the homework I gave that founder, and it is the most useful exercise in this whole post.

Write down two columns. What is comfortable about your current situation, and what is uncomfortable about it.

Then sort each column into good and bad. Some comfort is earned and healthy. Some comfort is a cage with a nice chair in it. Some discomfort is a signal you are growing. Some discomfort is just a bad situation you should leave.

Four buckets. Good comfortable. Bad comfortable. Good uncomfortable. Bad uncomfortable.

Most people go their entire lives without ever auditing their relationship with comfort and discomfort. They feel it and they react to it, and they never look at it directly.

Once you have the four buckets written down, hand them to AI and ask it what it notices. Ask it which comforts are the most expensive. Ask it what a version of you twelve months ahead would move out of the bad comfortable column first.

This works because you supplied the raw material. The audit is yours. The pattern recognition is the machine's.

What Do You Do After AI Names Your Blind Spot?

You change the sentence, and then you build evidence for the new one.

Changing the sentence is fast and it matters more than it sounds like it should. "I am that number" becomes "I weigh that number." "I am hitting a wall" becomes "I have not found the next move yet." "I am not good at this" becomes "I am learning this." That founder made the last swap himself mid sentence on our call without noticing, which is how I knew he already had the capacity for it. The identity you are running on is built out of sentences you repeat, and you speak things into existence whether you meant to or not.

Building evidence is the second half, and it is the half people skip.

Ask AI this: based on everything you know about me, where am I already behaving like the version of me I say I want to be, and not giving myself credit for it?

The answers will surprise you. This founder had recently walked a hundred thousand steps in a single stretch. That is not the behavior of the person he was describing himself as. He already had the evidence. He had filed it under something else.

Then give the new identity somewhere to show up tomorrow. Pick one small repeatable action and run it for twenty one days. I use a one mile a day rule for exactly this. The distance is not the point. Doing it on the day it is inconvenient is the point, because that is the day the identity actually installs.

What to Actually Do This Week

1. Put your real context somewhere AI can read it. Goals, history, values, numbers, and transcripts of you thinking out loud.
2. Run the prompt. "Knowing what you know about me, tell me what you see. Tell me the hidden and non-obvious that I may be hiding."
3. Ask it to go one layer deeper. Take the second answer seriously.
4. Do the comfort audit. Four buckets, written by hand, then handed to AI for pattern recognition.
5. Rewrite the one sentence you now know you have been repeating.
6. Ask AI where you are already living the new version and not crediting yourself.
7. Pick one action and run it for twenty one days.

The Real Reason This Works

Every time you say "I do not know how," that is a signal to put it into AI. Every time something makes you uncomfortable, that is a signal it is probably the right thing. Those two triggers, used consistently, will move you further than any tool you add this year.

What stands in the way becomes the way. The thing you keep walking past is the thing.

The founder in this story did not learn anything new about himself on that call. He heard something he had already said, out loud, at a volume he could not ignore. AI got there first because he gave it enough of himself to work with.

Give it enough of yourself to work with. Then ask it the hard question and sit with the answer.