The Number Most Executives Won’t Say Out Loud

There is a number I have not been able to put down.

Ninety-one percent of senior executives quietly admit they pretend to understand AI better than they actually do.¹

Read that again, because the size of it is the whole point. This is not a fringe. It is almost everyone in the room. The person across the boardroom table nodding along when someone says "agentic." The peer who posted about their AI rollout. The board member asking you pointed questions. Most of them are performing a fluency they do not feel.

If you are somewhere in that ninety-one percent, I want you to hear this before anything else: you are not behind, and you are not alone. You are just the one who has to stand in front of the room with the answers while the ground keeps shifting under everyone's feet.

The quiet performance nobody admits to

You have gotten good at the nod. The one you give when a term comes up that you were supposed to have absorbed by now, while you make a private note to look it up later.

You have learned to say "we're exploring that" in a way that sounds like a strategy instead of a stall.

And when you open the tool yourself, late, after the calls are done, you type something in, get something back that is fine, and quietly wonder if this is really the thing everyone keeps insisting will change your business.

None of that makes you a fraud. It makes you honest with yourself, which is rarer than it sounds. The research bears out how common the private version of this is: as many as half of employees now hide their own AI use from the people they work with, unsure whether being seen using it, or being seen not knowing how, will count for or against them.² The whole organization is performing confidence at each other. Leaders most of all, because leaders have the least room to wobble in public.

Here is the part that reconciles the older number with this year's data. Ask executives directly, and their confidence looks like the opposite of a confession. In the 2026 State of AI for Business report, senior leaders were the most confident group in the entire survey. CEOs and founders were the only role where a majority rated their ability to evaluate AI as high, well ahead of their own directors, vice presidents, and managers.³ Both things are true at once. The public confidence is the performance. The 91% is what sits underneath it. And it makes sense the gap is widest at the top, because the higher you sit, the more fluency everyone assumes you already have, and the less room you have to say otherwise.

Even the leaders who are genuinely ahead feel it. In that same 2026 research, someone considered an advanced user wrote that they feel like they are falling behind every day. A leader put it more plainly still: finding the time to lead, learn, and experiment is hard.⁴ That is not the language of people who have this figured out. It is the language of people carrying it quietly. And these came from an unusually AI-forward crowd, which means most leaders are further back still.

I spend a lot of my weeks sitting with leaders like you. And when the door closes and the performing can stop, what I hear is almost always the same quiet sentence. I don't actually know if I'm using this well.

That sentence is not a weakness. It is the beginning of getting good at this.

Why it makes complete sense that you're here

Look at what you have been handed.

A technology that changes faster than any briefing can keep up with. A news cycle that swings between "this replaces everyone" and "this is a bubble," sometimes in the same week. A market full of people selling certainty about a thing none of them fully understand either. And a board that reasonably wants to know what you are doing about all of it.

Of course you feel behind. The conditions were built to make anyone feel behind. Feeling lost in this moment is not a sign that you are slow. It is a sign that you are paying attention.

So let me take the shame out of the room entirely, because it was never useful anyway, and put something more useful in its place: an honest map of where you actually are, and where this can go.

The real gap isn't knowledge. It's altitude.

Almost no one says this part plainly.

Most leaders are not using a weak version of AI. They are using a tiny corner of a strong one. They are treating a capable strategic partner like a search box with better manners. They ask it a question, get an answer, close the window, and conclude that the tool is fine but oversold.

The gap is not really about knowledge. You are standing on the ground floor of something with a lot of floors, and no one has shown you the staircase.

I think about this as a climb from chatbot to coworker. On the ground floor, AI answers your questions. At the top, it does real work alongside you, holds the context of your business, challenges your thinking, and moves your goals forward between conversations. Almost everyone starts on the ground floor. Almost no one is told there is anywhere else to go.

So here is the staircase.

Where are you, really? The climb from chatbot to coworker

Read these six and notice, without judgment, where you honestly sit. Most leaders are on the first rung and calling it done. That is not a failing. It is just a starting line that no one drew for you.

1. Speak to it like you'd brief your sharpest new hire. The difference between "make this better" and giving it the real context, the goal you're after, and the constraints you're working inside. Better questions in, noticeably better answers out. This one change alone moves more than people expect.
2. Give it a memory. Build what I call a business brain: your context, your priorities, how you think, written down once so the tool stops meeting you cold every time. I keep a master context document for exactly this. When AI knows your business, it stops guessing and starts helping.
3. Plan with it before you turn it loose. Set the direction, tell it your goals, shape the approach together, and think out loud with it first. The best work happens when you spend your time planning alongside it, not firing off a single request and hoping.
4. Let it guide you. This is the wise-partner rung, and the one I lean on hardest. A good AI partner can hold your goals and milestones, keep the steps and measures that matter in front of you, and carry context forward so your best thinking doesn't die in a one-off session. I keep a tracking directive at the root of every project I run, so nothing important gets lost between conversations. Part sage partner, part quiet engine of progress.
5. Build your own board of directors. Stand up a few advisors inside the tool, personas that spar with you, pressure-test your logic, surface the blind spots and the biases you can't see in yourself. I do this in my own practice. It sits here, further up the climb, on purpose. Being challenged well takes trust first, and trust is something you build with a tool the same way you build it with a person.
6. Hand it real work. Connect it to the tools you already use so it doesn't just talk about the work, it does it. Drafts the report. Pulls the numbers. Prepares the brief. This is the top of the climb, the floor where "chatbot" finally becomes "coworker," and it is closer than it looks from the ground.

Read that list and one of two things is probably true. Either you are further down than you assumed, or you have quietly been doing more of this than you gave yourself credit for. Both are useful to know. Neither is anything to be embarrassed about.

You don't have to climb it alone

The loneliest place to not know something about AI is the corner office. It is the one seat where everyone assumes you already do.

I will tell you something I am not sure I am supposed to admit as the person writing this.

I did not learn any of these because someone handed me a framework. I learned them the way I learn most things, by sitting in the machine and doing the reps, one honest hour at a time, getting it wrong and adjusting. Ten years inside a growing company taught me that the leaders who get good at hard new things are almost never the ones who already understood them. They are the ones who found a safe place to not know, and a partner who would not make them feel small for asking.

That is the version of this work I care about most right now. I am not talking about another course you will half-watch. I mean a real conversation about where you actually are with AI, in plain language, with no one keeping score, and a way to move up the climb that fits how you actually lead.

If any of this landed

If you recognized yourself somewhere in that ninety-one percent, the most useful first step is not to study harder. It is to say the quiet sentence out loud to someone who has heard it before.

I would be glad to be that person. A discovery call is free, and it is exactly what it sounds like: a conversation, not a pitch, to figure out where you are and whether I am the right partner to help you get further. If you would rather locate yourself privately first, sit with the six rungs above and answer honestly which floor you are standing on. That answer, on its own, is worth more than most of what you will read about AI this month.

You already know more than you are letting yourself believe. And the part you don't know yet is far more learnable than the noise around you wants you to think.

That is the work. And it starts the moment you stop performing and start asking.

References

1. Pluralsight. "Pluralsight Research Finds that 79% of Tech Workers Pretend to Know More About AI Than They Actually Do." Press release, April 2, 2025. Survey of 1,200 technology decision-makers and practitioners in the US and UK; 91% of C-suite executives reported pretending to know more about AI than they do. pluralsight.com
2. Laserfiche. "Nearly Half of Employees Hide Workplace AI Use, Pointing to a Need for Openness and Policy Clarity." 2025 workplace AI survey. Estimates vary across 2025 studies, from roughly one-third to more than half, on the share of employees who conceal their AI use from managers or colleagues. laserfiche.com
3. SmarterX. 2026 State of AI for Business Report. Survey of 2,109 respondents, February–April 2026. 49% rated their confidence in evaluating AI as high or very high (up, year over year, from roughly 43%); CEOs, founders, and presidents led at 63%, the only role above 50%, ahead of directors (46%), other C-level (45%), VPs (42%), and managers (42%). Note: the survey audience skews AI-forward, so broader mid-market confidence likely runs lower. smarterx.ai
4. SmarterX, 2026 State of AI for Business Report, open-ended participant responses. Verbatim comments from respondents describing their biggest current struggle with AI.