The gap between what executives say about AI and what they actually do with it is widening. Public endorsements, strategy decks, and innovation committees abound. Quietly, implementation stalls, tools get purchased and shelved, and eventually, pilots launch and disappear. Nazma M. Rosado, an executive coach and change leadership strategist who works with organizations navigating digital transformation, has a clear view of what is driving that disconnect, and it has nothing to do with technology.
“It’s not going to go away,” Rosado says. “We need to know what to leverage and how to leverage it. Leaders need to stop, think, and strategize before implementing.” The hesitation is not irrational. The cost of getting AI wrong: compliance failures, eroded trust, and workforce resistance is real. What is irrational is letting that hesitation become permanent inaction while competitors move forward.
Build Personal Fluency Before Organizational Momentum
The instinct for most CEOs is to start big. Rosado inverts that entirely. The first move is personal, not organizational. Use AI to support your own decision-making, run research through it, and test it on low-stakes problems in a private environment before introducing it into professional workflows. “Try it out in a safe space so that you’re comfortable with it before you launch it into your daily work practice,” she says. “Use it to do some of your mundane personal tasks until you feel comfortable with it. I always advise my clients to practice with low stakes and implement with full confidence.”
A concrete starting point she recommends is to stop assigning someone to take meeting minutes. Record the meeting, let AI produce the transcript and summary, and keep everyone fully present in the conversation. It’s simple, immediate, and reversible if it does not work. “Not only can you utilize this to save time, but you can also pop that transcript into an AI tool to assess how effective the meeting was, and what can be improved.”
The logic behind starting small is strategic, not cautious. A CEO who has never personally used AI cannot credibly lead an organization through adopting it. Fluency precedes authority. “Start slow, gain your footing, then gradually increase, and practice what you preach,” Rosado says. Leaders who skip this step and push AI adoption from a distance create quiet resistance from the people they need on board.
Trust Is the First Casualty of a Rushed Rollout
Speed is where most AI implementations break. When organizations move faster than their people can process the change, employees stop trusting the outputs, start fearing for their roles, and revert to familiar methods. The tools are in place. Nobody is using them honestly. “If you go too quickly, people stop believing the outputs,” Rosado says. “They start fearing for their roles, fearing how much value they’re contributing versus how much is being replaced. And they quietly resist.”
The solution is not slower technology deployment. It is deliberate change management running in parallel. This involves clear parameters around what AI is being used for, room for employees to experiment and fail safely, and honest communication about what changes and what does not. On the question of job displacement, Rosado is direct. AI will eliminate the transactional elements of certain roles. That is not a threat. It is the point. “It frees up your time and ability to think, strategize, and drive innovative thinking to move the company forward.”
The Leadership Shift Nobody Is Preparing For
As autonomous AI agents take on functional roles within teams, leadership itself has to evolve. Managing effort and output is no longer sufficient. The real work becomes managing judgment, how decisions connect, how reasoning is applied, and where human accountability begins and ends. “Leaders now need to design their goals to be much more definitive, provide guardrails, and establish accountability,” Rosado says. “What are we expecting our internal people to do versus AI agents? It should be a combined execution, not one or the other.”
Leadership that once focused purely on people, then expanded to include processes and systems, now has to integrate machine learning into that same network. Influencing ethics, critical interpretation of AI outputs, and holding space for the kind of deep strategic judgment that no model can replicate; these are the leadership capabilities that will define organizational performance over the next decade.
“We need to speed up on the things that are transactional and slow us down,” Rosado says, “while holding that space for deeper thinking about how this impacts the world at large.” That balance, between acceleration and reflection, is where the real work of leadership lives right now.
Follow Nazma M. Rosado on LinkedIn for more insights on AI adoption, change leadership, and building organizations that thrive through transformation.
