AI is transforming not only how we work, but also how we lead. While AI can increasingly handle data analysis, pattern recognition, and decision-making at superhuman speed, core leadership capabilities remain deeply human—and are more critical than ever.

Put simply, machines stumble where empathy, ethics, and intuition hold sway—that is, in almost any situation where people work together toward a shared goal. Future machines may well console, judge, and invent on par with humans, but they’re not there yet. Rather than hanging up your leadership hat, embrace AI as a way to extend, not diminish, your leadership impact.

Smart leaders harness the strengths of AI while doubling down on the human capabilities that cannot be automated.

Two reasons why AI can amplify—but not replace—great leaders

1. Machines optimize—people innovate

AI excels at crunching data to refine strategies and identify opportunities. True breakthroughs, however, require questioning assumptions and embracing uncertainty—domains where today’s algorithms predictably stall. Leaders who treat AI as a copilot, not an autopilot, will unlock the best results.

2. People’s opinions matter—a lot

In an age of data overload, trusted relationships matter more than ever. People change because of people—especially friends, family, and peers—not because of data alone. We rely on trusted others to signal which information deserves our attention. Leaders who cultivate psychological safety and invite open, candid exchange unlock the full power of human ingenuity.

Five uniquely human leadership superpowers

Empathy: AI can generate an “empathetic” response, but it lacks nuance—and people notice. Empathy voiced by AI often feels flat, inauthentic, and uninspiring. AI empathy may fool some of the people some of the time, but not for long. Sensing the unspoken fears and aspirations in a room remains a distinctly human strength.

Ethical judgment: Weighing moral tradeoffs in ambiguous situations—where data alone leads to brittle decisions—remains the realm of human decision-makers. AI lacks innate moral reasoning and may propose “solutions” that ignore ethical boundaries. Because these systems learn from historical data, they can perpetuate or even magnify unethical practices. For now, it is up to human leaders and developers to design and apply the right guardrails for AI.

Intuitive leaps: Human judgment remains critical for connecting disparate ideas into novel strategies for escaping the “local optimum” plateau that can trap algorithms. Humans are subject to cognitive biases, but we can learn to recognize and counteract bias in ways generative AI cannot.

Artificial General Intelligence (AGI) is meant to close this gap. Depending on whom you ask, AGI could be just around the corner or still a decade away. If and when it arrives, AGI may be able to innovate on its own. For now, however, AI is constrained to recombining what already exists. Of course, human innovation often does exactly that—as Sir Isaac Newton put it, by standing on the shoulders of giants—but not always.

Resilience: Rallying people through setbacks requires authentic vulnerability, not scripted responses. AI cannot compete with the “Ted Lasso effect”—knowing when to challenge, when to encourage, and when to buy them a pint. Humans have evolved to navigate complex social environments and form nuanced psychological, social, and organizational responses to adversity. AI has none of this lived experience.

Storytelling: AI‑generated stories can be engaging, but research shows they often lack the depth, authenticity, and relational impact of human‑crafted narratives. Readers report feeling less “transported” into AI‑generated story worlds than into human ones. (As an aside, the thought‑provoking novel Playground by Richard Powers takes on just this question.)

Leadership Actions

Use AI to free up people for high-touch work

Map your processes end to end and identify tasks AI can automate, such as reporting or initial triage. Done well, this can reclaim 20–30% of your team’s time for relationship‑building and creative problem‑solving. With automation, however, comes new risk. Use a project pre-mortem to address that risk: gather your team, assume your AI‑supported initiative has failed, and explore what could have caused it. Surface risks such as data bias or skill atrophy, then design safeguards like mandatory reviews or skill development.

Build in friction to reduce risk

MIT’s Renée Richardson Gosline warns against an overreliance on efficiency gains from AI without considering the long-term costs that may come from a “frictionless” environment. AI removes effort, but when things are too easy, we lose our capabilities in critical thinking and problem solving. This is what I call the Navi effect—as we increasingly rely on our car’s navigation to “get us there,” we may lose the ability to find the way home on our own. The antidote is to build in friction and deliberately design for thoughtful engagement and human oversight.

Train human–AI “hybrids” with ethical guardrails

Pair each team member with a tailored AI tool (for example, an analytics copilot or content generator) and train them explicitly on ethics and guardrails. Agree upfront on non-negotiable principles—such as transparent decision criteria and “do no harm”—and then test adherence through peer review of outcomes and processes.

Redesign incentives for synergy—and host AI free offsites

Measure and reward success on hybrid outcomes such as “insights actioned” or “team innovations launched,” with bonuses tied to thoughtful, collaborative use of AI. Schedule regular AI free offsites for unstructured idea jamming to foster intuitive leaps and deeper connection. Host “human edge” feedback sessions to recognize and celebrate non-AI contributions from team members.

The bottom line: Amplify, don’t automate

AI is poised to raise the bar for great leaders and their organizations. It can increasingly handle the “what” and the “how,” but people still own the “why” and the “so what.”

The real question is not whether AI will lead.
It is how you will lead people in a world shaped by AI.

Reflection questions

Which of my decisions demands irreplaceable human judgment?

How am I using AI to amplify, not dilute, my team’s strengths?

Where might over reliance on AI trap us in a local optimum.

Who can I rely on to help navigate AI’s ethical gray zones?

How do I create space for human intuition amid the data deluge?

References

“I, ChatGPT: Linguistic properties and human experiences of human- versus AI-generated stories.” (2025). Humanities and Social Sciences Communications. https://www.nature.com/articles/s41599-025-06341-2​

Andon, P. (2025, December 17). In praise of friction: Why the future of AI needs more resistance. BusinessThink, UNSW Business School. https://www.businessthink.unsw.edu.au/articles/ai-positive-friction-productivity-human-oversight

Ethical leadership in the age of AI: Challenges, opportunities and strategies. (2024). arXiv. https://arxiv.org/html/2410.18095v2​

Generative AI lacks the human creativity to achieve scientific discoveries. (2025). Scientific Reports. https://www.nature.com/articles/s41598-025-93794-9​

Harvard Business School. (2025). AI won’t make the call: Why human judgment still drives innovation. https://www.hbs.edu/bigs/artificial-intelligence-human-jugment-drives-innovation​

Human resilience in the AI era: What machines can’t replace. (2025). arXiv. https://arxiv.org/abs/2510.25218​

Lawrence, N. D. (2024). The atomic human: Our place in the age of artificial intelligence. Cambridge University Press.

Marr, B. (2024, May 8). The important difference between generative AI and AGI. Forbes. https://www.forbes.com/sites/bernardmarr/2024/05/08/the-important-difference-between-generative-ai-and-agi/

McKinsey & Company. (2025, January 27). Superagency in the workplace: Empowering people to unlock AI’s full potential at work. McKinsey & Company.

Milotich, M. (2025). Breaking through the local optimum trap. LeadingWell with Mark Milotich. https://markmilotich.substack.com/p/breaking-through-the-local-optimum​

Milotich, M. (2025). Facts don’t move people—peers do. LeadingWell with Mark Milotich. https://markmilotich.substack.com/p/facts-dont-move-peoplepeers-do​

Rubin, M., et al. (2025). Comparing the value of perceived human versus AI-generated empathy. Nature Human Behaviour. https://pubmed.ncbi.nlm.nih.gov/40588597/​

Seth, N. (2025). Human edge in the AI age: Eight timeless mantras for success. Bloomsbury Publishing. ​

Spiegelhalter, D. (2024). The art of uncertainty: How to navigate chance, ignorance, risk, and luck. Cambridge University Press.

The HOW Institute for Society. (2025). AI and ethical leadership. https://thehowinstitute.org/ai-and-ethical-leadership/