When Institutions Get Smarter, Leaders Can’t Stay the Same

Davos has a very particular feeling that’s hard to explain until you’ve been there.

It’s the crunch of snow under boots at 7 a.m., the sharp cold that wakes you up faster than coffee, the low hum of conversation spilling out of hotel lobbies and temporary pavilions before the sun fully clears the mountains. It’s overhearing debates about geopolitics, energy systems, and AI governance while standing in line for a kebab and realizing the person next to you might be shaping policy, capital flows, or global infrastructure.

And this year, I heard one conversation over and over. 

Across sessions on AI, workforce transformation, and risk, a consistent tension kept surfacing: institutions are becoming more adaptive, more automated, and more responsive—while many leaders are still relying on decision habits shaped for a slower, more linear environment.

AI is changing how decisions are formed, how quickly they move, and how many people and systems influence the outcome. 

For many of the leaders I work with, the question underneath all the strategy discussions is deeply personal: “What does this require of me now?”

Smarter Systems, Familiar Leadership Friction

In a session hosted by Workday, focused on closing the AI workforce gap, one statistic caused a noticeable pause in the room: roughly 40% of AI-generated output requires rework, largely because organizations are deploying AI without adjusting the human decision structures around it.

This is where leadership friction begins to show.

I see this same pattern in my work. Companies invest heavily in tools while leaving judgment models, accountability boundaries, and decision authority untouched. The result is faster output paired with uncertainty about ownership, quality, and responsibility.

Research from MIT Sloan reinforces what many leaders are experiencing firsthand. Human–AI collaboration underperforms when leaders don’t recalibrate how people evaluate, challenge, and take responsibility for machine-generated outputs. AI tends to amplify whatever leadership system already exists.

From Gross Output to Decision Quality

Aashna Kircher, Group general manager and CHRO of Workday, the importance of distinguishing between gross productivity and net productivity. Net productivity accounts for rework, review, exception handling, and the human effort required to validate outcomes.

That distinction matters because AI accelerates execution without resolving judgment.

Leaders are receiving cleaner, faster outputs, but they still have to decide what to trust, when to intervene, and how to explain outcomes to others. This is where static decision habits surface. Some leaders delay decisions in search of certainty that never arrives. Others move too quickly without clarifying accountability.

Here’s the one contrast worth naming clearly: AI doesn’t replace judgment; it intensifies the consequences of weak judgment.

Decision Velocity As A Leadership Stress Test

AI compresses time. Decisions that once unfolded over weeks now occur in days or hours. Leadership development, however, still largely assumes linear timelines and extended deliberation. And under these conditions, leaders fall back on defaults.

Research on executive decision-making under uncertainty consistently shows that leaders perform best when they have internal frameworks for judgment rather than relying on instinct, hierarchy, or avoidance. Without those frameworks, speed magnifies bias and stress responses.

This is why AI adoption often feels psychologically heavier than technically complex. Leaders are being asked to decide faster, with higher visibility, and with fewer buffers.

Flattening Hierarchy, Rising Expectations

Another theme running through Davos conversations was the erosion of rigid hierarchy. Career ladders are dissolving and functional silos are blending. AI accelerates this shift by routing work based on context rather than titles.

Authority, however, doesn’t disappear—it changes form.

Leaders are increasingly expected to explain decisions, justify tradeoffs, and create clarity without relying on positional power. Credibility now rests on steadiness, transparency, and the ability to communicate under pressure.

This aligns closely with what I’ve written about executive presence—not as polish, but as judgment. Presence today is demonstrated by how leaders hold ambiguity and explain decisions when certainty is unavailable.

Where Leaders Actually Get Stuck

Across these conversations, leadership challenges clustered around a consistent set of issues:

• Difficulty making decisions when AI outputs appear confident but incomplete

• Blurred accountability when systems influence outcomes

• Increased pressure from faster decision cycles

• Psychological resistance rooted in fear rather than capability

• Governance treated as secondary rather than foundational

• 
  

None of these originate in the technology itself. They emerge when leadership systems haven’t evolved alongside it.

When Models Meet Human Reality

One observation that stayed with me was the reminder that models eventually collide with reality.

AI adoption carries real psychological weight. Fear of loss—status, relevance, control—often shapes behavior more than technical readiness. Organizational research consistently shows that resistance is less about tools and more about uncertainty and perceived risk.

Leaders who acknowledge this dynamic openly tend to reduce friction. Leaders who ignore it inadvertently create it. I’ve written before about how the instinct to tighten control often masks discomfort rather than strength and AI brings that instinct to the surface quickly.

What This Means for Leaders Now

All of this points to a clear conclusion: institutions are adapting faster than many leaders are updating how they decide.

The leaders who navigate this well are not distinguished by technical fluency. They are distinguished by their ability to hold accountability when systems are involved, communicate tradeoffs clearly, and make decisions that withstand scrutiny.

This is the work I do with leaders. Strengthening judgment, sharpening presence under pressure, and preparing executives for decision environments that are faster, less linear, and more exposed.

As institutions grow smarter, leadership shifts away from control toward judgment, clarity, and responsibility. AI requires leaders to evolve and the leaders who update how they decide—not just what they deploy—will shape what intelligent institutions become.