For fourteen months our archive has been saying the same thing.
AI governance is not a tool problem. It is a structure problem. The people at the top have not caught up with the people doing the work.
Microsoft just published the evidence.
Twenty thousand workers. Ten countries. Trillions of productivity signals from inside Microsoft 365. The findings landed this week and one number stands above the rest.
Organizational factors — leadership, culture, manager support — have double the AI impact of individual factors like employee mindset and behavior.
Double.
Read that again slowly. Because it means everything that follows.
Workers are not the problem.
Fifty-eight percent of AI users are producing work they could not do a year ago. Without being asked. Without a roadmap handed down from the executive floor. Without a policy document or a committee or a consultant brought in at four hundred dollars an hour to explain what a prompt is.
They figured it out. They experimented on their lunch break. They built new habits in the margins of a workday that was already full. They adapted because they understood — at a gut level, before the memos arrived — that the tools were changing and standing still was not an option.
The people on the ground did their part.
Their leaders did not follow.
Only one in four AI users believes their leadership is clearly and consistently aligned on AI strategy.
One in four.
That means three out of every four people trying to work effectively with AI are doing it without a coherent signal from the top. They are navigating without a map because the people who are supposed to draw the map are still debating whether they need one.
The rest are caught in what Microsoft calls a transformation paradox. Afraid to fall behind if they don’t adapt. Afraid to adapt because the organization has not changed the incentives, the metrics, or the meeting structure to support it. Just thirteen percent report being rewarded for using AI to reinvent how they work.
Think about what that number means in practice.
A person figures out how to do in two hours what used to take two days. They share it with their team. Nobody above them notices. Their annual review still measures the old outputs on the old timeline. The innovation happened and disappeared without acknowledgment.
That is not a technology problem. That is a leadership failure with a very specific shape.
Here is what that failure costs in hard numbers.
The organizations that actually redesigned their operating model around AI — not bolted AI onto old systems, not added a chatbot to the existing process, but rebuilt how work gets done from the structure up — saw eighty percent of their people producing work they could not do a year ago.
The organizations still waiting for leadership to catch up are sitting at fifty-eight percent.
Twenty-two points.
That gap is not a product gap. It is not a training gap. It is not a budget gap. It is a leadership gap, measured in output, sitting in plain sight in a study that nobody at the top of most organizations will read carefully enough to feel uncomfortable.
The pattern is not new.
Leadership has been late to every major technology shift of the last thirty years. The internet arrived and organizations spent a decade arguing about whether they needed a website. Mobile arrived and companies built desktop experiences and wondered why engagement dropped. Social arrived and the communications department wrote a policy about it while the workforce was already three platforms ahead.
AI is not different. The response to it is identical.
The tools land. The workers adapt. Leadership watches. Then leadership commissions a study. Then leadership schedules a meeting about the study. Then leadership forms a working group to develop a framework based on the meeting about the study.
By the time the framework is ratified the workers have moved on to the next thing.
This is the cycle. Microsoft has now documented it with twenty thousand data points across ten countries and the conclusion is the same one anyone paying attention already reached.
The bottleneck is not capability. It is not willingness. It is not the technology.
It is the people in the room who are supposed to be leading and have decided that watching is close enough to leading to count.
One honest note before we close. This is a Microsoft study. The executive quoted most heavily runs Copilot and Agents at Microsoft. The findings are credible. The methodology is serious. The source has a product interest in the conclusion and you should know that when you weigh it.
It does not change what the numbers say.
The real problem is older than AI and it will outlast this news cycle.
You cannot govern what you have not structured. A policy document sitting in a shared drive while the organization runs on the same habits it had in 2019 is not governance. It is decoration. It is the appearance of leadership without the weight of it.
The Baseline was built on that argument before Microsoft had the data to prove it.
Now they do.
The question for every leader reading this is a simple one.
Are you the one in four. Or are you the reason your people are working without a map.
“The Faust Baseline Codex 3.5”
”AI Baseline Governance”
Post Library – Intelligent People Assume Nothing
“Your Pathway to a Better AI Experence”
Unauthorized commercial use prohibited. © 2026 The Faust Baseline LLC
