Leadership in the AI Age Starts with Trust

Published: May 15, 2026 | By Jim Hutcherson

Students at the University of Central Florida recently pushed back hard against a commencement speech that praised AI. The reaction spread quickly because it captured something many leaders are still underestimating. Resistance to AI is not showing up as a technical issue. It is emotional, cultural, ethical, and personal.

Most employees already know AI can summarize a meeting faster. What concerns them is whether their experience still matters as AI takes on more of their work. Many are trying to understand whether leadership has already made a decision about their future.

For years, the IT shift was treated mostly as an operational problem. The pattern was familiar. Design a new process, roll out the platform, train users, measure adoption, then move to the next phase. Many AI programs are still following that same model, even though AI behaves differently once it reaches the daily work.

The difference shows up once the tool starts influencing decisions, and people have to determine whether the answer can be trusted.

AI touches judgment, confidence, status, and professional identity. It reaches into the part of work that people associate with their value and competence. When the reaction is that personal, adoption numbers can hide more issues than they can reveal. I have learned to be careful with these numbers, as adoption can look cleaner than it is. A login does not mean someone trusts the answer. A completed prompt does not mean the team knows what to do with the result. The report may show activity while the team is still manually checking the answer, asking a trusted coworker, or avoiding the decision the tool was supposed to support.

I am seeing more AI launch problems where the rollout assumptions break down before the technology does. The misalignment usually shows up early. Leadership sees that the initiative has started because the demos look good and the first reports show movement. What is not seen is that the team is still deciding whether the tool can be trusted. Then usage flattens. People attend the meetings, nod through the presentations, and continue doing the work the old way. Employees learn how to perform AI enthusiasm just enough to stay out of trouble.

Organizations often view training as the root cause of transformation issues. The cause is more likely trust.

The organizations moving fastest with AI rarely have the most advanced tools. Typically, these are the ones where people can experiment without feeling exposed. A manager who says, “Run a small test and show us what you learn,” usually gets a different response than a manager who announces rollout targets before the team has enough experience with the tool to trust the process.

Some resistance comes from fear of exposure. People wonder what happens if they use the tool wrong, if the machine writes a better first draft, or if everyone can see they are not moving as fast as expected. Ethics is a different conversation. When people ask whether the tool is fair, whether the data should be used, or whether a decision can be trusted, they may be seeing a real risk rather than simply resisting change. The two conversations sound similar on the surface, but they usually need different responses.

None of this is entirely new. People have always needed room to say, “I do not understand this yet,” or “I tried it, and it did not work,” without paying a price for being honest. Amy Edmondson’s research on psychological safety made that point years before anyone was talking about ChatGPT at work (Edmondson, 1999). Newer AI research is reaching similar conclusions from a different direction. People appear more willing to engage with AI when trust already exists inside the team (Reich et al., 2026).

The leaders handling this well stay close to the work. They listen after the meeting, not just during the presentation. They notice when a team double-checks an answer, pauses before using it, or brings in someone with experience before making the decision. Those moments tell you more than the dashboard.

The more grounded conversations tend to stay close to the tasks. What changed in the decision? What became easier? What almost got missed? Where did experience still matter? The discussion becomes less about whether AI is good or bad and more about how judgment shifts within the process.

Technology adoption always looks cleaner on a presentation slide than it feels inside a working team. Real organizations move at the speed people can absorb change while still doing the work in front of them. When leaders move faster than the organization can process, resistance often goes underground and shows up later as disengagement, workarounds, or avoidance.

AI can accelerate analysis, compress research time, generate first drafts, summarize meetings, and recommend next steps. It cannot sit with a team after the demo, hear what people are worried about, and decide what needs clarification before the next step. That part of the transition still depends on people.

Start with one task.

One pattern continues to stand out. Teams seem to learn faster when the work stays concrete. A single task often reveals more than a broad strategy discussion because people can see what improved, what became riskier, and where experience still carries the decision. Speed often shows up first. Capability is what compounds.

References

Bremen, J. (2026, February 17). Overcoming barriers to AI adoption in 2026. Forbes. https://www.forbes.com/sites/johnbremen/2026/02/17/overcoming-barriers-to-ai-adoption-in-2026/

Edmondson, A. (1999). Psychological safety and learning behavior in work teams. Administrative Science Quarterly, 44(2), 350-383. https://www.jstor.org/stable/2666999

Fortune. (2026, April 16). AI resistance is running out of time as rebellion, quiet quitting, and trust issues reshape the workplace. Fortune. https://fortune.com/2026/04/16/ai-resistance-running-out-of-time-rebellion-quiet-quitting-trust/

Hermann, E., Puntoni, S., & Morewedge, C. K. (2026, March). Why gen AI feels so threatening to workers, and what leaders can do to ease the anxiety. Harvard Business Review. https://hbr.org/2026/03/why-gen-ai-feels-so-threatening-to-workers

Orlando Weekly. (2026). UCF students boo commencement speaker over AI praise. Orlando Weekly. https://www.orlandoweekly.com/news/orlando-area-news/ucf-students-boo-commencement-speaker-over-ai-praise/

Reich, A., Wolfe, D., Price, M., Choe, A., Kidd, F., & Wagner, H. (2026). Safety first: Psychological safety as the key to AI transformation. arXiv. https://arxiv.org/abs/2602.23279

ScienceDirect. (2024). Leadership, organizational adaptation, and AI adoption research. ScienceDirect. https://www.sciencedirect.com/science/article/pii/S1053482224000652

Tampa Bay Times. (2026, May 12). UCF commencement speaker met with boos over AI remarks during graduation ceremony. Tampa Bay Times. https://www.tampabay.com/news/education/2026/05/12/ucf-florida-commencement-speaker-ai-nvidia/

TechRadar Pro. (2026). 2026 is the year CEOs must rewire the C-suite. IBM study reveals what successful leaders are actually doing with AI in their businesses. TechRadar Pro. https://www.techradar.com/pro/2026-is-the-year-ceos-must-rewire-the-c-suite-ibm-study-reveals-what-successful-leaders-are-actually-doing-with-ai-work-in-their-businesses/