The mirror and the spark

There is a quiet irony running through much of the AI conversation in leadership circles. We are using the most sophisticated pattern-recognition technology in human history to help people become more self-aware. And yet, self-awareness (real self-awareness) is precisely what AI cannot produce. 

That gap is worth sitting with. 

For coaches, executives, and HR leaders navigating the integration of AI into development work, this is not primarily a technology question. It is a developmental one. And how we answer it will shape whether AI becomes a genuine accelerant to human growth, or an elaborate shortcut that leaves people more informed and no more transformed. 

 

A polarity, not a problem 

The arrival of AI in coaching and leadership development presents what Barry Johnson, author of Polarity Management, would recognize immediately as a polarity: two interdependent values that cannot be resolved by choosing one over the other. Machine Intelligence and Human Presence are not competing options. They are, in Johnson’s terms, the inhale and the exhale. 

Treat them as a problem to solve (pick AI or pick the human) and you end up in the swamp of one extreme or the other. Manage them as a polarity, and you access something neither can provide alone. 

This distinction matters enormously for practitioners. Most of the current conversation about AI in coaching is operating at the level of a dilemma: Will AI replace coaches? Should we use it or resist it? These are the wrong questions. The more generative question is: what does each pole offer, where does each become a liability, and what becomes possible when we hold both? 

 

What machine intelligence offers 

AI is, at its best, a remarkable feedback instrument. It surfaces patterns without the friction of a human relationship. It holds what you’ve said across time without forgetting, without projecting, without needing anything back from you. It is patient in a way humans structurally cannot be. 

For a leader embedded in the noise of organizational life (the meetings, the politics, the endless demand to perform) that kind of clean reflection has real value. It offloads cognitive weight. It flags the gap between stated values and actual behavior. It can do in seconds what a coach might spend several sessions excavating. 

This is what I’d call the utility pole: AI as a mental concierge, clearing the analytical brush so that something more essential can emerge. 

But here is the developmental trap. When leaders stay in the mirror, they accumulate insight without integration. They become expert observers of their own patterns, articulate about their blind spots, fluent in the language of self-awareness. And none of it actually changes how they show up under pressure. The map becomes a substitute for the territory. Development stalls at the level of content. 

The danger is not that AI knows too little about us. The danger is that we mistake its knowledge for our own.

 

What human presence actually provides 

The coach, the trusted peer, the mentor who has watched you across a decade. These relationships carry something algorithms cannot model: the capacity to hold a person’s becoming, not just their current state. 

What skilled coaches and HR leaders bring is not primarily knowledge or even skill. It is structure of mind. The ability to hold the complexity of a person without collapsing it into a framework or rushing it toward resolution. To notice what isn’t being said. To stay present with ambiguity long enough for something genuine to surface. 

This is what American developmental psychologist, Robert Kegan, means when he distinguishes between the content of our thinking and the structure that shapes it. AI can work with content extraordinarily well. It cannot shift structure. That requires a different kind of encounter, one that involves genuine risk, genuine relationship, and genuine otherness. 

The shadow of this pole is equally real. Without data, without pattern, without the clarity a good instrument can provide, human development work can drift into echo chambers of intuition. Coaches get attached to their own frameworks. Executives get lost in narrative. HR leaders, exhausted by the relational weight of the work, start burning out on the very thing they’re trying to cultivate in others. 

 

Consciousness as the return on investment 

The organizations getting this right are not choosing between AI efficiency and human depth. They are using AI to reclaim bandwidth, and then investing that bandwidth somewhere the machine cannot go. 

This is the question that should be driving every HR leader and coaching practice right now: when we free up attention through intelligent technology, what do we do with it? If the answer is ‘more meetings’ or ‘more output,’ we have simply offloaded one treadmill onto another. 

The developmental case for AI is this: when we use machine intelligence for what it does well (pattern recognition, reflection, information synthesis) we create the conditions for humans to do what only humans can do. To be present. To take genuine risks in relationship. To ask the questions that don’t have algorithmic answers: Who are we becoming? What do we actually stand for? What would it mean to lead from here? 

In the Cook-Greuter and Torbert lineage, this is the movement from self-awareness as a cognitive achievement to self-awareness as a mode of being. AI can accelerate the former. It has nothing to offer the latter. That leap still requires the full weight of human encounter. 

The machine can show you the threads. Only you can decide what they’re worth weaving together. 

 

What this means in practice 

For coaches: the question is not whether to integrate AI tools but how to position them developmentally. Used well, they can surface material faster, create richer pre-session data, and help clients sustain reflection between sessions. Used poorly, they become a way for clients to stay safely in their heads, informed but unchanged. 

For executives: the temptation will be to use AI’s clarity as a substitute for the discomfort of genuine development. Pattern recognition without relational risk is not growth. It is optimization. Both have their place. Only one of them changes who you are. 

For HR leaders: the strategic question is how you design development ecosystems that use AI for acceleration and human relationships for transformation, without collapsing the two or treating them as interchangeable. This requires developmental literacy, not just technological fluency. 

AI is an extraordinary mirror. Let it do that work. Let it show you the patterns with a clarity that human egos rarely can. 

And then put the mirror down. Turn toward the people in the room. Bring the full, inefficient, irreplaceable weight of your presence. That is where development actually happens. Everything else is rehearsal.