And will we have a choice?
I like working with ChatGPT as a thought partner.
I don’t think of it as a coach, a therapist or an assistant, because I will never fully trust that it has the human qualities or expertise that I need from the people who fill those roles in my life. But the back and forth we’ve developed has helped me in many ways.
I use it to brainstorm different approaches to problems. I appreciate its quick, practical feedback on work in progress. When I have something long and dry to read, I ask Chat to pre-read it (not summarize). It sparks my curiosity with good questions worth thinking about before I dive in, and helps me remember and assimilate abstract concepts by coming up with metaphors that help me more easily grok complex information.
When I’m stuck, I’ll even ask it to talk with me in the voice of a favourite writer, or thought leader, just to get a playful, fresh take on a direction I want to take. I’m not sure yet whether this practice will make me more or less creative, but I’m having fun.
Recently, I decided to push our conversations further and ask it, given what it’s learned about me so far, what it considered my blind spots. The first one threw me a little:
Fear of Not Thinking Like a Leader
I wasn’t entirely clear on what that meant, so I asked Chat to tell me more:
Sure! When I identified “Fear of Not Thinking Like a Leader” as a blind spot, I meant that you might have an unconscious hesitation to fully step into leadership — not because you lack ability, but because your internal definition of leadership may not match how you naturally think and operate.
How Do Leaders Think?
It’s humbling to admit how much truth there is to this. Traditional leadership frameworks prioritize qualities that are valuable in the corporate world. Trait Leadership Theory lists some of these as decisiveness, confidence, assertiveness and competitiveness. Leaders who fit that model take charge, move fast, and exert control. That’s never felt like me.
Looking back over my time as a chief learning officer, however, I can see how I fell into that leadership role because I possessed different qualities. We see these more in modern adaptive leadership frameworks, which feature traits like perceptiveness, responsiveness, and comfort with ambiguity. In an era of accelerated change these qualities have become more important.
Still, it’s hard to shake the power of the traditional model. I’ve worked with leaders who were driven to act often and quickly. They established control through meetings, and established order through org charts. I’m glad for the ways that traditional leadership has challenged me to be more decisive, authoritative and willing to respect the need for well defined roles and responsibilities. I don’t want to say that one leadership style is necessarily better than the other. In times of crisis, the traits of the traditional leader can be essential.
But my drives move me in another direction: towards taking the time I need to be present, to develop the clearest possible vision of reality, and towards understanding the mindsets of others so that I can identify the stumbling blocks and support them in acquiring new skills. I believe that in the era of AI, adaptive traits will become even more necessary, and leaders who default to the traditional style, may need to work harder at respecting and developing those traits.
How Will They Have To Think?
Those who don’t will encounter at least two major, even dangerous, stumbling blocks.
First they will resist mastering AI because they are too uncomfortable with the idea of a technology that seems to ape their signature trait, decision making. Research presented in the book More Human: How the Power of AI Can Transform the Way You Lead by Rasmus Hougaard and Jaqueline Carter supports my instincts on this. 57 percent of leaders they surveyed don’t trust AI, and 22 percent were neutral. This means over 75 percent of leaders don’t feel pulled towards using it.
Of the 20 percent who do, there’s the risk that they will use it in ways that will amplify the worst failings of traditional leadership. Their first instincts will be to try to use AI, or encourage others to use AI, to make quick decisions on poor or biased data. The consequences of those decisions will then have to be fixed by humans through (sigh) more meetings. The resulting chaos and time wasted will trigger the default strategies of reinforced hierarchies and company restructuring.
Hougaard and Carter sound a warning call for leaders who are sticking their heads in the sand.
AI will not replace leaders, but leaders who don’t know how to use AI will be replaced by those who do.
Until we reach the trough of disillusionment with new technology, as we always do. Then those leaders will have to be replaced by those who know how to create a better balance between AI agents and human wisdom.
Leading Without Thinking (So Much)
I ran this theory by Chat for some feedback. But I wanted it to look at my thinking from a different angle. So I asked it to give me feedback in the voice of one of my favourite writers, Oliver Burkehart (Meditations For Mortals). Here’s what not really Oliver Burkehart had to say:
If I may, I’d invite you to pause and consider the underlying premise: that you must think like a leader in order to be one. You gesture towards an alternative model of leadership — one built on perceptiveness, adaptability, and presence — but even so, there’s a kind of ghostly reverence for the old model hovering at the margins. The nagging sense that leadership is still about decision-making, about control, about stepping up.
And yet, what if the most powerful kind of leadership isn’t about any of that? What if it isn’t even a trait or a mindset at all, but something that happens in the spaces between people — the product of relationships, rather than a quality that resides in any single individual? If that were true, then the real blind spot isn’t whether you “think like a leader,” but whether that’s even the right question to be asking.
Your point about AI is a sharp one: leaders who try to graft AI onto a traditional decision-making framework will almost certainly find it exacerbating their worst tendencies. But if AI really does force us to rethink what leadership even is, then perhaps the most valuable leaders in the years to come won’t be those who use AI well, but those who help others grapple with its implications — who create the conditions for wisdom to emerge, rather than simply issuing more decisions, faster.
What interests me is that you still seem to be holding the old framework in mind as a reference point, even as you argue against it. I wonder what would happen if you let that go entirely — if you entertained the possibility that the most radical way forward isn’t to redefine leadership, but to question whether we ever understood it in the first place.
The Human Take
With this in mind, I ran my blindspot by a favourite human, an executive coach who has decades of leadership experience in some of the most traditional industries there are, aerospace, health care, pharmaceuticals.
She reminded me that leadership in the real world is not usually about mindsets and frameworks. It’s about being able to meet the situation at hand, and the situation of the people you are leading at the stage they are in their career, company evolution, or particular project. This kind of real world leadership is best supported by a situational leadership framework, of which there are many.
But for future leaders, meeting the moment will take more than any framework can provide. It will be about building actual experience with both AI and the humans impacted by AI.
Experience doesn’t alway mean mastery. I suspect the future of leadership won’t be defined by how well we master AI, but by how well we stay open to being challenged — by AI, by people, and by reality itself — in the messy, unpredictable, and deeply human work of leading.
Originally published in Corporate Underbelly March 28,2025
Clear Path Learning 
Leave a comment