We Redesigned Leadership and Forgot to Tell the Leaders

When I started interviewing leaders for The Leadership Transition Report, I assumed most of our conversations would revolve around AI itself. I expected discussions about tools, governance, adoption, productivity and use cases. I thought we’d spend our time talking about technology.

Instead, I spent months talking to leaders whose jobs no longer felt like the jobs they’d signed up for.

I remember one executive pausing halfway through our conversation and saying, “I’m not finding AI difficult. I’m finding everything around AI difficult.” He wasn’t talking about prompts or models. He was talking about trying to hit quarterly targets while simultaneously changing how 300 people thought about work.

The more conversations I had, the more I realized people weren’t struggling to understand AI. They were struggling to understand what leadership now required of them because AI had arrived.

That distinction matters.

Across different industries, different leadership teams and organizations at completely different stages of AI adoption, the same themes kept surfacing. Work feels heavier, decisions feel more complicated, and there are more competing priorities than there used to be. The expectations of the role have expanded, yet almost nothing has been taken away.

If you’re a leader reading this, I imagine some of that feels familiar.

Your quarterly targets haven’t changed. Your customers still expect the same service. Your team still needs coaching. Difficult conversations haven’t disappeared. Budgets still matter. Hiring still matters. Performance still matters. Culture still matters.

None of that has gone away. Instead, another layer has been added on top.

Now you’re expected to identify where AI creates value, encourage your team to experiment, make judgment calls about when AI is useful and when it isn’t, rethink long-established ways of working and help people adapt to changes that many of them are still trying to make sense of themselves.

At the same time, you’re expected to answer questions that don’t have clear answers.

“How should I be using AI?”

“Will this change my role?”

“Should I trust this output?”

“What’s the right balance between doing it myself and letting AI help?”

Leadership has always involved uncertainty, but today many leaders are being asked to create confidence while navigating uncertainty themselves. That is a very different job.

One of the clearest gaps I found was how much investment organizations are making in teaching people how to use AI, and how little is going into helping leaders think differently because AI exists. Those aren’t the same thing. Learning how to write a better prompt is useful. Learning how to lead a team whose way of working is fundamentally changing is something else entirely.

Where are we helping leaders develop the judgment to know when AI should inform a decision and when experience, context and human understanding matter more?

Where are we helping managers navigate conversations with employees who are worried about what AI means for their future when nobody can honestly guarantee what every role will look like five years from now?

Where are we helping leaders build confidence in people without pretending to have certainty themselves?

Those aren’t technology questions. They’re leadership questions.

Another observation stayed with me long after I finished the report.

For years, many organizations rewarded leaders for becoming experts. The deeper your functional knowledge, the more valuable you became. Success often meant knowing more than anyone else in your area. That’s not enough anymore. AI is rewarding something different - curiosity, connection, and the ability to see patterns across functions rather than optimize one of them.

The leaders I see adapting best aren’t necessarily the ones with the deepest expertise. They’re the ones learning to think across the organization rather than only within it. They’re becoming more curious about how work connects between teams, asking different questions, and getting more comfortable making decisions without having every answer first. In other words, they’re thinking more horizontally than vertically.

That shift sounds subtle. It isn’t.

AI doesn’t respect organizational boundaries. It allows people in Finance to do work that previously sat with Data. It gives Marketing access to capabilities that once belonged to Engineering. HR can automate work that previously took weeks. Operations can analyze information that once required specialist support.

The knock-on effect is that leaders can no longer think only about their own function. They have to understand how work flows across the organization because that’s increasingly how AI works too.

There was one final theme that I couldn’t stop thinking about.

Every team contains people with different levels of confidence around technology. Some embrace every new tool immediately. Others are more cautious. Some may never become power users, yet they bring years of commercial judgment, customer insight or industry experience that no AI model can replicate.

Leadership now means creating an environment where both groups succeed. If we only celebrate the fastest adopters, we risk losing diversity of thought. If we ignore AI altogether, we risk falling behind. Holding those two ideas in tension has become part of the leadership role, and nobody taught most managers how to do that.

Looking back, the biggest lesson I took from writing The Leadership Transition Report wasn’t about AI. It was about leadership.

I don’t think leaders have suddenly become less capable than they were two years ago. I think we’ve fundamentally redesigned the role. We’ve asked leaders to coach differently, think differently, communicate differently, make decisions differently and lead people through a level of ambiguity that many have never experienced before. At the same time, we’ve continued to measure success against expectations that belonged to the previous version of the job.

That’s why so many leaders I’ve spoken to feel like work has become heavier. Not because AI is making leadership harder, but because we redesigned what leadership means, and we’re still developing leaders as though nothing has changed.

If any of this feels familiar, the full report goes deeper - eight patterns I kept finding in organization after organization, regardless of sector or how far along the AI rollout was.