AI Is Making Leaders Busier, but Not Necessarily Better

Artificial intelligence is rewriting what it means to lead well. Not because it threatens to replace leaders, but because it is quietly sorting them into two camps: those who use it to sharpen their thinking, and those who unknowingly let it do their thinking for them.

The stakes are higher than most leadership teams realize.

When AI entered the workplace, many predicted a relaxation dividend. Intelligent machines would handle the tedious work, and professionals, including leaders, would finally have breathing room to think strategically. That has not happened.

Research from ActivTrak, analyzing the digital behavior of over 10,000 employees, found that AI adoption intensified work rather than easing it. Time spent on email, messaging, and business software spiked dramatically among early adopters, who squeezed work into previously unproductive hours and took on tasks they had previously delegated. A parallel finding from researchers at UC Berkeley's Haas School of Business confirmed the pattern: AI makes tasks tractable enough that workers expand their scope rather than their leisure.

For leaders, this creates a paradox. AI gives you more capacity, but capacity for what? The answer to that question will define leadership careers in the decade ahead.

Intelligence Is No Longer the Differentiator

For most of organizational history, leadership advancement tracked closely with cognitive horsepower. The smartest people in the room tended to rise. AI complicates this calculus fundamentally. When raw analytical capability becomes abundant and affordable, and when any knowledge worker can summon a first draft, a market analysis, or a legal summary in seconds, intelligence alone stops being a competitive moat.

What takes its place?

Psychologists have long studied a trait called need for cognition, meaning the degree to which individuals genuinely enjoy effortful thinking. People high in this trait seek out hard problems, persist through ambiguity, and find intellectual struggle intrinsically rewarding. Research led by John Cacioppo at the University of Chicago, reviewing more than 100 studies on the trait, found that high-need-for-cognition individuals engage more deeply in complex reasoning, form stronger convictions, and pursue problems with greater persistence. Their drive comes from within, not from external pressure or deadlines.

In a world where AI can approximate the outputs of average analytical effort, the leaders who will differentiate themselves are those who bring something AI cannot replicate: genuine hunger to understand, decide well, and take ownership of conclusions.

Volition, the will to engage, is becoming the scarcest leadership resource.

Three Leadership Patterns Emerging Right Now

Look closely at how leaders are currently responding to AI, and three patterns emerge.

The Delegation Leaders. Some leaders treat AI as a senior staff member to whom hard thinking can be outsourced. They generate AI-drafted memos, AI-synthesized reports, and AI-produced strategy summaries, treating the outputs as their own conclusions. Productivity appears high. But neuroscience research suggests a hidden cost: when cognitive load is offloaded to machines, the brain adapts. A research team led by Nataliya Kosmyna at MIT's Media Lab found that brain connectivity declines by as much as 55 percent when people use ChatGPT compared with when they perform similar tasks unaided. Separately, Vivienne Ming, a co-founder of Possibility Sciences, found that gamma-wave activity, a neurological marker of active cognitive effort, dropped by roughly 40 percent when workers used AI to complete tasks.

Leaders who habitually outsource their thinking do not just produce lower-quality decisions; they gradually diminish their capacity to make good decisions at all.

The Resolute but Inconsistent. A second group understands the risk. They read the research, attend the workshops, and commit earnestly to using AI as a tool rather than a crutch. But the pull of optimization is relentless. Modern technology is engineered, as Google's former head of search once described it, to reduce every possible friction point between users and the information they want to find. Friction, it turns out, is where learning lives.

These leaders hold the line on high-stakes decisions but gradually cede smaller choices and routine communications to the algorithm. Over time, the cognitive muscle atrophies in ways they do not notice until a crisis demands it.

A study of physicians using AI in colonoscopy procedures, published in The Lancet Gastroenterology and Hepatology, found that after AI adoption, and then removal of the AI tool, detection rates for precancerous lesions dropped from 28.4 percent to 22.4 percent, because the habit of careful observation had weakened during the period of machine assistance. The same dynamic applies in the boardroom. Decision-making instincts degrade when they are not exercised.

The Augmentation Leaders. The third group uses AI the way master craftspeople use precision instruments, extending capability without replacing judgment. They bring a well-formed point of view before consulting the machine. They use AI to stress-test their thinking, surface perspectives they have not considered, and explore adjacent territory, not to generate the thinking itself.

These are leaders who have internalized what Carol Dweck's research on growth mindset established: the process of struggling with hard problems is itself what builds capability. They treat intellectual difficulty as a feature, not a bug. And they are increasingly outpacing peers who lean on AI for the heavy lifting.

The Productivity Trap: When the Tool Becomes the Time Sink

There is a subtler version of the busyness problem, and it has less to do with the quality of a leader's thinking than with the simple economics of a leader's time.

Because AI is so capable, it invites an almost bottomless stream of questions. You can ask it one more thing, then refine the answer, then chase a tangent, then refine again. Each step feels productive. Taken together, they can quietly become a black hole for hours that were meant for higher-value work. The tool is genuinely useful, which is exactly what makes the drift so easy to miss.

The same trap appears when leaders turn to automation. AI can now take over recurring tasks, and that is a real gift. But there is a temptation to sit with the machine and instruct it, correct it, and train it to perform a specific task exactly the way you want. That effort can consume an enormous amount of time. In practice, it is often a disguised form of doing the work yourself. The stronger move is usually to hand the high-level task to someone else and let them do the work of building the automation with AI.

The principle is simple, and it is easy to lose sight of. Just because AI can do something for you does not mean you should be the one spending the time to make it do it. Some work genuinely requires your judgment and your presence. Other work should be delegated, whether to a person, to the machine, or to a person working with the machine. Confusing the two is how capable leaders end up busy and behind.

AI is the shiny new tool, and shiny tools invite play. The discipline that separates strong leaders is the willingness to use it strategically, with clear boundaries on how much of their own time it deserves, rather than getting pulled into tinkering because the tinkering feels like progress.

The Strategic Risk Leaders Aren't Talking About

There is a team-level version of this risk that most organizations are ignoring entirely.

When leaders reward AI-accelerated output without asking how that output was produced, are they inadvertently training their teams to produce impressive-looking work that is brittle under scrutiny? This is a leadership failure hiding behind a productivity story.

The most forward-thinking leaders are drawing deliberate lines. They are establishing norms around where AI judgment is acceptable and where human judgment is non-negotiable. They are creating protected time for unassisted analysis. They are asking their teams not just what they found, but how they reasoned through it.

What AI Cannot Do, and Why That Is a Leadership Advantage

Beneath the productivity question lies a more fundamental one about what leadership actually is.

AI systems predict. They synthesize. They optimize toward defined objectives with impressive efficiency. What they cannot do is want something: to feel the weight of a decision, to care about outcomes beyond the immediate prompt, or to grow from failure in a way that shapes future choices.

Leadership, at its core, is not primarily computational. The judgment that matters most in organizations, including the call to pursue an uncertain market, to hold a value when it is costly, to develop a person who does not look ready yet, emerges from a leader's accumulated experience, their personal history of bets made and lessons absorbed, and their particular sense of what is worth fighting for.

These are qualities that deepen through practice and atrophy through disuse. AI does not make them irrelevant. It makes them rarer, and therefore more valuable.

A Framework for Leading Well in the AI Age

The leaders who thrive in this environment will not be the ones who resist AI or the ones who surrender to it. They will be the ones who build deliberate habits around it:

1. Form the view before you ask the machine. Before consulting AI on any significant question, spend time developing your own analysis. AI's role should be to challenge your thinking, not originate it.

2. Protect the hard conversations. Performance reviews, strategic pivots, and difficult feedback require presence and judgment that no prompt can simulate. Guard them accordingly.

3. Build AI literacy into your team culture. Great leaders create space for their teams to discuss not just what AI produced, but how to evaluate, challenge, and override it. Critical evaluation of AI outputs is a skill that must be taught.

4. Model cognitive effort. When leaders visibly engage with difficult problems, working through complexity rather than delegating it to algorithms, they signal that depth matters in their organization. Culture follows what leaders demonstrate, not what they declare.

5. Invest in apprenticeship. Research consistently shows that learning accelerates through structured mentorship, being guided by someone who models not just competency but the disposition to work hard, stay curious, and persist through difficulty. In an AI-saturated environment, this kind of human transfer of craft becomes more valuable, not less.

The leaders who will matter most in the years ahead are not necessarily those with the most access to the best AI tools. They are the ones who maintain the will to do the hard cognitive work that makes AI useful, bringing enough of themselves to the partnership that the machine amplifies something real.

That has always been what distinguished great leadership. AI just makes it harder to fake.