What Your Next Staff Meeting Says About AI Transformation

Companies have spent the last three years building AI pilots, governance structures, transformation roadmaps, and dashboards. The investment is visible. The progress is not always obvious.

According to McKinsey, 6% of organizations qualify as AI high performers, meaning they attribute at least 5% of their earnings to AI use. Forty-four percent have scaled AI across the organization, up from 38% a year earlier. The tools are present. The capability is not evenly distributed.

The reasons are rarely technical. According to the Microsoft Work Trend Index, 65% of employees fear falling behind if they do not adopt AI, but only 13% feel rewarded for experimenting with it. Microsoft calls this the Transformation Paradox. Organizational factors, including culture, leadership support, and talent development, have twice the impact on AI success compared to individual effort.

That gap shows up in the most ordinary place in a company: the staff meeting. The questions leaders ask, the topics they reward, and the behaviors they reinforce week to week are a better indicator of transformation than any dashboard. If the meeting is still about milestones and delivery, then AI is a project. If it becomes about what was learned, what surprised the team, and what will change, then AI is becoming part of how the organization works.

Shelly Swanback has spent more than three decades leading technology and business transformation inside large organizations, including through earlier waves of digital change. Her perspective on AI comes from having seen organizations confront this challenge before. New technology arrives quickly, but changing the way people lead, learn, and work takes much longer.

For Swanback, that history matters because many of the challenges companies now associate with AI are not actually new. The technology may have changed, but the fundamental questions around how organizations adapt have not. Her argument is direct: AI is not simply a technology problem. It is a people, leadership, and culture problem.

"We learned a lot of this back in the digital era," Swanback says. "The lessons still apply. Sometimes the simplest things work best. It does not all have to be brand new, fancy, and complicated."

What she means by simple is specific. In her own experience as a CEO, she changed the questions her leadership team asked. Swanback states, "Instead of only asking whether projects were on track, I added three questions: What did we learn? What surprised us? What will we do differently?" The shift moved the conversation from delivery to adaptability.

"The whole notion of AI is that it always needs to be learning. But so does your organization. For the organization to adapt and actually move things forward, it has to learn as fast as the AI is learning," she adds.

That creates a different standard for what AI transformation looks like. The measure is not simply how quickly a company deploys new tools, but how quickly the organization can absorb what those tools reveal and adjust around them. A pilot can produce useful information in weeks, but if decisions, behaviors, and ways of working remain unchanged months later, the organization has not kept pace with the technology it introduced.

Swanback describes the failure mode as treating learning as a side activity. Many organizations have a team that is good at experimentation and continuous learning. That team often sits in the innovation function, and the rest of the organization does not operate that way. Learning is an annual exercise, a required training, and a workshop. It is not part of the work.

"Organizations want to learn. They want to be adaptable," Swanback says. "But most still do not make learning part of the work. It is still a side thing."

The practical implication is that transformation does not require a new initiative. It requires changing what happens in the meetings that already exist. Swanback describes a practice she used as CEO: inviting someone who, on paper, did not need to be there. A finance person might join a marketing meeting, for example, not to add oversight but to bring a different perspective and help learning move across functions.

The same principle can be applied in smaller ways. Add one question about learning to a status report, or highlight an insight from another part of the organization during a global staff meeting.

None of these changes are dramatic on their own. That is precisely the point. Transformation becomes visible when experimentation and learning stop being exceptional activities and begin showing up in the ordinary rhythms of the organization: the questions asked in a Monday meeting, the perspectives invited into the room, and what leaders choose to recognize and carry forward.

The question for leaders is not whether they have an AI strategy. It is whether the ordinary conversations in their organization have changed. If they have not, the strategy is not working yet.