AI accelerates insight and analysis, but effective leadership still requires thoughtful human insight.

AI moves fast, but strong leadership keeps humans in the loop to reflect and guide decisions.

An opportunity exists to combine AI’s speed with humans in the loop for better judgment, ethics, and results.

Artificial intelligence can summarize research, analyze data, generate ideas, and propose strategic decisions in seconds. It’s a true breakthrough.

But when it comes to leadership, it’s never been just about simple analysis to get to the right decision. It’s also about making sense of complexity combined with uncertainty, weighing trade-offs, understanding people and their motivations, and considering long-term consequences.

This creates a paradox at the center of modern leadership: AI encourages speed, but great leadership often requires slowing down to reflect. And the leaders who succeed in today’s AI-frenzied world are likely to be those who know when, and how, to do both.

The Hidden Risk of AI-Driven Decisions

AI adoption is happening fast. According to a survey conducted by McKinsey last year, nearly 90 percent of the companies surveyed reported experimenting with AI.

AI is already being used support human reflection and personal insight. Many leaders are now using AI tools as thinking partners. Instead of asking AI only for answers, many leaders now use it to explore questions like:

What assumptions might I be making in this situation?

What perspectives might I be overlooking?

What risks could emerge from this decision?

Applied this way, AI could help expand thinking.

But using AI doesn’t automatically make you a great leader. In fact, many of the leaders I’ve worked with are feeling a bit overwhelmed by how to fully implement and benefit from AI. If anything, AI has challenged their leadership versus made it better.

Research shows that people often rely on mental shortcuts when making decisions, especially under pressure. Daniel Kahneman’s work on “System 1” and “System 2” thinking explains the difference. System 1 thinking is fast and intuitive. System 2 thinking is slower and more reflective. In fast-moving organizations, leaders often default to System 1 thinking because they just don’t feel like they have the time to slow down.

AI can unintentionally reinforce this pattern by delivering quick answers that feel authoritative.

One concept that has gained some traction in leadership research is the idea of the “human loop.” The idea is simple: AI systems are most effective when they augment human decision-making rather than replace it.

Researchers Elizabeth Graswich and Jennifer Sparks Taylor, authors of The Human Loop: Leading with Reflection in the Age of AI, outline how leaders should reflect on multiple dimensions of a decision, challenge, or opportunity before acting:

Know yourself: Understand your own values, experiences, and biases so you can interpret AI insights thoughtfully and ensure decisions align with what you believe is right, not just what the data implies.

Know your people: Consider how AI-driven decisions affect employees, customers, and stakeholders, using empathy and awareness to ensure technology supports people’s goals rather than simply replacing them.

Know your environment: Reflect on the broader organizational, cultural, and regulatory context so AI fits the realities of your industry, company culture, and societal expectations.

Know your impact: Think through the short- and long-term consequences of AI-enabled decisions, including how they influence trust, fairness, relationships, and organizational reputation.

Know your self-awareness: Continuously reflect on your assumptions and decisions, using both personal reflection and tools like AI to challenge your thinking and improve your judgment.

The goal is not to slow organizations down unnecessarily but ensure that fast insights are balanced with thoughtful, reflective judgment.

Examples of Humans in the Loop

Many successful AI implementations include humans in the loop. At the Mayo Clinic, for example, nurses use AI to draft responses to patient messages. Previously, nurses were inundated by patient communications. Today, AI receives and crafts the initial response but nurses still review and edit the messages before they are sent.

Nurses now save an average of 30 seconds per message. And the communications are often more detailed and empathetic than what a time-pressed nurse could write from scratch. It’s a win-win-win for nurses, the organization, and patients.

Another example comes from Microsoft. CEO Satya Nadella has often described how raising a son with cerebral palsy profoundly changed how he sees the world. That experience deepened his sense of empathy and shaped his leadership philosophy. Rather than viewing technology purely through the lens of efficiency or performance, Nadella began to emphasize how innovation should improve people’s lives—especially for those who face daily barriers.

That perspective influenced Microsoft’s AI strategy. The company launched initiatives like AI for Accessibility, which funds technologies designed to empower people with disabilities, and developed tools such as Eye-Gaze, which allows individuals to type using only eye movements. In this case, personal reflection at the leadership level helped ensure that AI innovation focused not just on what technology can do, but on how it can better serve human needs.

In both cases, the technology drove innovation. But thoughtful leadership and application ensured AI served real human needs.

The Leadership Questions That Matter Most

As AI becomes more integrated into organizations, leaders will increasingly face new kinds of decisions. Leaders need to move beyond just asking “Can we automate this?” and explore questions like:

Should we automate this?

Who benefits from this technology and how?

What unintended consequences could emerge?

What are our ultimate goals?

These questions slow the decision process. They create space for leaders to reflect on impact, ethics, and long-term outcomes—and that reflection can lead to better decisions.

Go Slower to Go Faster

The biggest mistake organizations can make with AI is assuming speed is the goal. The faster technology becomes, the more valuable thoughtful reflection is likely to become, too. The emerging paradox of leadership is this: Those who combine AI-driven insights with deliberate reflection could create better strategies, build more trusting teams, and anticipate risks that others overlook. Basically, you need to go slower to go faster.

As AI expands across organizations and society, the real leadership advantage will come from knowing when to move at machine speed, and when to keep humans in the loop.

Graswich, E., & Sparks Taylor, J. (2025). The Human Loop: Leading with Reflection in the Age of AI. The Human Loop.

Kahneman, D. (2011). Thinking, fast and slow. Farrar, Straus and Giroux.

Javed, M., Pless, N., Waldman, D. A., Garavan, T., Gull, A. A., Akhtar, M. W., Mouri, N., Sengupta, A., & Maak, T. (2024). What, when, and how of responsible leadership: Taking stock of eighteen years of research and a future agenda. Journal of Management Studies. Advance online publication. https://doi.org/10.1111/joms.13157

McKinsey & Company. (2025, November 5). The state of AI: How organizations are rewiring to capture value. McKinsey & Company. mckinsey.com/capabilities/quantumblack/our-insights/the-state-of-ai

Kaplan, S. (2023). Experiential Intelligence: Harness the Power of Experience for Personal and Business Breakthroughs. Matt Holt Books.

© Psychology Today