The Most Important Leadership Skill Might Be Unlearning

Most of us think about learning as adding something.

A new skill. A new idea. A new perspective. A new way of working.

But sometimes learning requires something much harder:

Letting go of something we already think we know may be the best leadership skill.

That may be particularly important right now.

AI is rapidly changing what’s possible at work. Tasks that once took hours can now be done in minutes. Work that required specialized knowledge can increasingly be supported by technology. Information can be analyzed, created, summarized, and shared in different ways.

Yet there’s a risk that organizations will introduce all this new technology while leaving most of their old assumptions about work untouched.

We add AI to the existing process.

We automate individual tasks.

We make today’s work a little faster.

Those things can create value.

But there’s a much bigger question leaders should be asking:

If we were designing this work today, with everything that’s now possible, would we design it the same way?

Increasingly, the answer may be no.

Organizations Accumulate Assumptions

Every organization operates on assumptions.

Some are explicit. Many aren’t.

We assume existing roles, decision structures, processes, meetings, expertise, and customer needs are fixed simply because they have been in the past.

None of these assumptions is necessarily wrong.

That’s what makes them difficult to see.

Most organizational assumptions exist for a reason. At some point, they probably made sense.

But organizations change. Customers change. Technology changes. People change.

And right now, AI is changing some of the fundamental economics and possibilities of how work gets done.

The danger isn’t having assumptions.

The danger is continuing to operate on them after the conditions that made them true have changed.

AI Should Make Us Question More Than Technology

A lot of the conversation about AI starts with tools.

Which platform should we use, what tasks can we automate, how much time can we save, and where can we boost productivity?

But they can also keep us anchored to today’s organization.

Imagine taking an existing ten-step process and using AI to make three of the steps faster.

That’s improvement.

But what if, with today’s technology, we wouldn’t design a ten-step process at all?

That’s a very different conversation.

The bigger opportunity with AI may not simply be doing today’s work faster.

It may be giving us permission to rethink how the work should happen in the first place.

And that requires leaders to approach the future differently.

Not from:

How can technology improve what we already do?

But from:

Knowing what’s possible now, what would we design differently?

That’s where unlearning becomes important.

Unlearning Doesn’t Mean Forgetting What You Know

Experience matters.

Organizations shouldn’t discard everything they’ve learned simply because a new technology arrives.

And leaders shouldn’t abandon practices that work just because something new is possible.

Unlearning is something different.

It’s being willing to examine whether what worked before still makes sense now.

It’s moving from:

This is how we do it.

to:

Why do we do it this way?

And eventually:

If we were starting today, would we still do it this way?

That shift sounds simple.

In practice, it can be difficult.

Because the more successful something has been in the past, the easier it is to assume it should continue into the future.

Experience tells us what worked before. Curiosity helps us determine whether it will work again.

A Simple Unlearning Test to Build Leadership Skill

As organizations think about their future, leaders might apply three questions to the assumptions that shape how work happens today.

1. IDENTIFY: What Are We Assuming?

Start by making the assumption visible.

Listen for phrases like:

We can’t do that.

We’ve tried that before.

Our customers expect this.

That role has always done that.

That’s just how this works.

Then ask:

What are we assuming to be true?

The goal isn’t to challenge everything for the sake of challenging it.

It’s to recognize when an old assumption is quietly determining a future decision.

2. TEST: Is It Still True?

Once the assumption is visible, test it.

What evidence tells us it’s still valid?

What’s changed since we first made that assumption?

What does AI or another technology make possible today that wasn’t possible before?

What are customers or employees doing differently?

And perhaps most importantly:

What would we expect to see if our assumption were wrong?

That last question matters.

It’s easy to find evidence supporting something we already believe.

It’s much harder, and often much more valuable, to deliberately look for evidence that challenges it.

3. REPLACE: What Makes More Sense Now?

Challenging an old assumption isn’t enough.

If it no longer serves us, what should replace it?

Maybe a ten-step process becomes three.

Maybe work moves between people and AI differently.

Maybe a role changes because some tasks disappear while entirely new responsibilities emerge.

Maybe decisions move closer to the people doing the work.

Maybe a meeting disappears.

Maybe expertise gets organized differently.

Maybe leaders spend less time overseeing tasks and more time exercising judgment, developing people, and solving complex problems.

The objective isn’t simply to remove the old. It’s to create something better suited to what’s coming next.

Look at the Organization With Fresh Eyes

There are plenty of places leaders could apply this thinking.

Work: If we were designing this today, would we design it the same way?

Processes: Which steps exist because they’re necessary, and which exist because they’ve always existed?

Roles: If AI changes the tasks people perform, how should the role itself change?

Decisions: Are decisions happening where they make the most sense, or where they’ve historically been made?

Technology: Are we using AI to improve old ways of working, or to imagine better ones?

Leadership: What leadership behaviours helped us succeed in the past but may become less useful in the future?

The point isn’t to redesign everything.

It’s about stopping the assumption that everything needs to stay the same.

Going Back to School Sometimes Means Unlearning

Over the past couple of weeks, I’ve been playing with the idea of leaders and organizations going back to school.

I started with five questions leaders might revisit this fall.

Then I looked at all the things organizations teach people every day through what they reward, tolerate, and model.

There’s another lesson worth adding.

Preparing for the future isn’t only about learning what’s new.

Sometimes it’s about questioning what’s old.

AI will undoubtedly provide organizations with new tools, capabilities, and ways to improve productivity.

But the organizations that benefit most may not be the ones that apply new technology to the way they’ve always worked.

They may be the ones willing to look at their organization with fresh eyes and ask:

What do we believe to be true that might no longer be true?

And:

If we were starting today, knowing what we know now and what’s possible now, would we still do it this way?

If the answer is no, the next lesson might not be something you need to learn.

It might be something you need to unlearn.