The AI Divide — Skills, Ambiguity, and the Efficiency Paradox

Why do some leaders take to AI quickly and others stall? The standard explanations — age, technical comfort, change resistance — only get you part of the way.

This article walks through three observations that together explain more of what's actually going on. Each one stands alone. Taken together, they map the AI divide.

Skills compound — and waiting isn't as neutral as it feels

I'm not writing this to scare anyone or to persuade anyone. I'm writing it because I'm starting to feel a growing disconnect between how fast AI is evolving and how casually many of us, myself included at times, think about "keeping up."

If you're in a leadership role, especially one that depends on strategic thinking, judgment, and synthesis, where should you be in your AI learning journey right now?

Not long ago, feeling competent with AI meant being able to write a solid prompt, get a decent response, and move on. For a while, that felt like real progress. It probably was.

But over the last several months, something has shifted.

I've been talking with people who spend a lot of time experimenting with AI, and there's a shared, almost uneasy sense that the pace of change has accelerated. At the same time, in conversations with peers who use AI more lightly, I hear a growing confidence that they're "caught up."

Those two perspectives feel increasingly out of sync.

What's made me pause is the realization that AI skills seem to build on themselves. The value people are getting appears tied not just to the tools improving, but to the time they've already spent learning how to work with them. Practice and experience matter. And those gains don't reset when the technology advances. They compound.

That's left me wondering whether waiting is as neutral as it feels.

I don't have answers yet. What I do have is a growing curiosity about why capable, thoughtful leaders are having such different experiences with the same technology.

The divide is about comfort with ambiguity

That curiosity led me somewhere specific: I started looking for what was actually driving the divide between the leaders pulling ahead and the ones who felt like they were falling behind.

Some leaders feel AI has quietly and meaningfully changed how they work. Others, often just as thoughtful and experienced, have tried it and concluded it adds only minor value.

That contrast reflects what's often described as the AI divide. To be clear, this isn't about intelligence, seniority, strategic ability, or values and ethical positions on AI. Those are important conversations in their own right. What I'm describing here is something more practical: day-to-day use, real work, real decisions.

How can two people, using the same tools, with similar experience and judgment, walk away with such different conclusions?

One factor keeps surfacing for me: comfort with ambiguity.

For the last couple of decades, we've trained leaders to rely on deterministic tools. Think Excel. Put the right inputs in, and you get the right answer out. Every time.

AI doesn't work that way.

It's probabilistic. It generates nuanced responses. Sometimes it's impressive. Often it's generic or slightly off. For leaders who prize precision, that doesn't feel creative. It feels broken.

Imagine two leaders ask AI to draft an email explaining a new policy.

• Leader A reads the output and thinks, this isn't my voice. I'd have to rewrite this anyway. It would be faster to do it myself. They conclude AI isn't particularly useful for real work.

• Leader B reads the same output and thinks, the tone is off, but the structure is there. They iterate. A few minutes later, they have a usable draft and a better sense of how to guide the tool.

The difference isn't intelligence or skill. It's what happens when the first answer isn't quite right.

• Some people evaluate AI as a product. Is it good immediately?

• Others engage it as a process. Can I shape this?

That small difference compounds. Staying with the ambiguity early builds intuition, and that intuition makes future first attempts far more effective.

This is where many smart, efficient leaders quietly bounce off AI. They're used to tools that behave cleanly and predictably. When AI doesn't, it feels unproductive rather than promising.

The efficiency paradox

Comfort with ambiguity explains a lot. It doesn't, on its own, explain why some of the leaders who should be best positioned for AI — the high performers, the strategic thinkers, the ones who pick up new tools fast — are also some of the most likely to walk away.

"It's faster to just do it myself." If you've tried using AI for complex work, you've probably said this. And here's the thing. You're not wrong.

Highly effective leaders already have efficient ways of working. They know how to write, analyze, plan, and decide. When AI feels slower, more iterative, or slightly error-prone, it is labeled as not worth the effort.

This is the efficiency paradox.

Some of the most competent people are often the first to walk away from AI, because their baseline of efficiency is already so high. They judge it by a very reasonable question: Did this save me time right now?

The leaders who get the most out of AI tend to approach it differently. They lead with curiosity — about what the tool can do, where it struggles, and how it responds when pushed. Over time, that curiosity builds intuition.

They stop asking, "Is this faster?"

They start asking, "What can this help me think through?"

AI only becomes efficient after you've been inefficient with it for a while.

That tension reflects how high performers have been trained to work and why this shift can feel counterintuitive at first. What matters is that this early investment doesn't just affect today's output. It changes what future first attempts look like as the tools continue to evolve.

What this means for school leaders

These three observations — skills compound, ambiguity comfort matters more than intelligence, and the efficiency paradox traps high performers — describe the same gap from different angles. The gap isn't between leaders who "get" AI and those who don't. It's between leaders who engage with AI as a process that builds intuition over time, and leaders who judge it as a product that should already be working.

For school leaders, the implication is structural. The leaders most often shaping their school's AI posture are also the most likely, by background and habit, to be on the wrong side of that divide. The longer the gap stays open at the top, the harder it gets to close.

The work isn't just adopting AI. It's adopting it differently than you'd evaluate any other tool — with patience for the early inefficiency, comfort with ambiguous outputs, and the recognition that practice does, in fact, compound.