The Metacognition Gap: Using AI to Master Self-Awareness

An irreparable gap is beginning to widen between those who possess "metacognition" and those who do not.

Furthermore, the speed at which this gap is widening is faster than ever before. It might be more accurate to say that AI, acting as an amplifier, is exponentially pulling the two sides apart.

What prompted me to write this was encountering two contrasting episodes that were simply too different to ignore.

① The CEO who asks AI for "harsh feedback"

Recently, I had the chance to speak with a business owner I've known for a while.

This person is clearly brilliant, even from my perspective. You could say their business combat power is at a monster-like level. I've been indebted to them many times, and in our work together, I'm constantly struck by the incredible resolution of their thinking.

They told me they've been doing something interesting lately.

After taking AI minutes of a meeting, I plug them back into the AI and ask, "How much value did I provide in this meeting? Give me some harsh feedback." The feedback is quite harsh and can be discouraging, but it's proving to be great training.

That's what they said.

Apparently, the AI responds with things like this:

Your question at 2:23 PM was valuable as it shifted the discussion from point X to point Y.
On the other hand, this statement around 2:40 PM didn't contribute much to the main thread of the discussion.
In the latter half of the meeting, after 3:00 PM, you barely spoke. Was that due to a lack of catch-up on the topic, or were you hesitating to speak your mind? Which was it?
You have a tendency to spend too much time on fact-checking rather than moving the discussion forward.

They read this and then head into the next meeting.

Honestly, it gave me chills.

This person is already at such a high level, yet they are still trying to grow.

What's more, I was surprised to hear that they aren't just blindly dumping the minutes into the AI.

Their method involves first writing out their own self-perception: "In today's meeting, I think I demonstrated this strength, but I feel I failed in this area." They then submit that alongside the minutes.

When I instruct it to "check if my self-perception is accurate or off," I get feedback that is incredibly harsh but specific and tangible.

What does this mean?

It means they are using AI to verify the very "boundary" between what they understand and what they don't.

If you think about this structure, a tremendous loop begins to rotate:

Write out your own self-perception.
AI compares it with the minutes and points out discrepancies.
The gap becomes clear (= areas where metacognition was lacking are highlighted).
Next time, write the self-perception based on that gap.
Find a new gap...

The more this loop rotates, the smaller the gap between perception and reality becomes. In other words, the resolution of their metacognition keeps increasing.

Naturally, their behavior in meetings, the quality of their remarks, and their ability to drive discussions all improve.

...And a person who is already monstrously talented is running this loop multiple times every single week.

It's truly incredible.

② The person who says "I'm giving advice" vs. the person who says "I hate talking to them"

Chronologically, this happened a bit before the first story, but I had the chance to speak with another individual.

Let's call them Person A.

Person A is in a position to give advice to Person B.

Person A: The advisor
Person B: The advisee I happened to have opportunities to communicate with both A and B fairly often.

Person A (the advisor) told me this:

Yeah, it's tough looking after Person B lately, but I'm being very supportive and giving them advice. We're doing regular 1-on-1s too.

I replied with something like, "I see, that's great of you," and left it at that.

However, later on, I had a chance to talk to Person B (the advisee).

When I asked, "How have things been lately?" this was the response:

To be honest, I really hate the 1-on-1s with Person A. They aren't supportive at all, and they even yawn while I'm talking. I feel depressed from the moment I wake up on the days we have those meetings.

I was stunned.

Regarding the exact same 1-on-1 session, Person A perceives it as "being supportive," while Person B feels "that time is nothing but painful."

More than the size of this gap, what I found truly terrifying is the structure where Person A might never realize this discrepancy for the rest of their life.

Person A operates on the self-perception that "I am giving good advice."

As long as they hold that self-perception as a premise, Person A will never think to improve their 1-on-1 method—not even by a millimeter. Because they think they're already doing it well.

Since they don't think they need to improve, there's no way for their metacognitive abilities to grow.

And they remain unaware that Person B thinks they are "truly awful."

They aren't even aware that they aren't aware.

Unless someone steps up and tells them bluntly, "Hey, your 1-on-1 style is terrible," Person A will likely stay in this state forever.

And in most cases, no one is going to say that to your face.

You're lucky if you're still getting feedback.

What was the difference between the CEO and Person A?

When you look at the CEO from the beginning and Person A side-by-side, you see a difference in something more fundamental than skill or raw intelligence.

The CEO, despite being so successful, operates on the premise that "I am probably still not doing enough."

That's why they seek feedback from AI, and rather than just dumping data, they ask at a level of "My self-perception is this; is it correct?"

On the other hand, Person A operates on the premise that "I am doing fine."

Therefore, the very idea of going out to get feedback never even occurs to them.

This difference in "premises" manifests as a staggering divide over time.

The CEO is increasing the resolution of their metacognition every single week.

Person A is expanding and reproducing the gap between their perception and reality every single week.

It's terrifying to imagine how much of a difference there will be in one or two years.

Just for reference, here is an image I made for a note thumbnail to illustrate this concept.

Conditions for entering the virtuous cycle of metacognition

Now, here is the main point.

Based on these stories, I've thought about what is necessary to get onto that virtuous cycle of metacognition. I believe there are three main factors.

I've written about this in detail on note, so please take a look if you're interested.

note: The Metacognition Gap is Huge: A Memo to Avoid Being Left Behind