AI Is Expanding Employee Agency. Most Organizations Block It

Initially published on Forbes May 5, 2026

When you are promoted to a management position, you learn to manage through people. Your people take on the execution, and you gain the decision rights, ownership and control that come with the role.

But only if you manage people.

If what you are managing these days is a team of AI agents, something different happens. As AI agents reshape knowledge work, they take on growing portions of execution. But the decision rights, the ownership, the authority to direct and shape outcomes — none of that transfers automatically. AI is expanding what can be described as human agency, but you are still measured on what you do, not on how you decide, direct or own.

Now scale that dynamic across an entire organization.

That gap has a name: “blocked agency.”

Microsoft’s 2026 Work Trend Index, based on a survey of 20,000 workers across 10 countries and analysis of trillions of Microsoft 365 productivity signals, puts precise data behind this shift. As AI takes on execution at work, humans should gain what the report calls agency: the ability to direct the work, make the calls, and own the outcomes. The urgency for every organization is to turn that expanded agency into value.

Most are not doing it. And the reason is not that people are not ready.

Here is what the data reveals about why organizations are blocking the agency AI is handing their people, and what leaders can do about it.

The AI Capability Gap: What Employees Can Do vs. What Organizations Allow

When employees work with AI, something shifts in what they are capable of. They no longer need to spend the bulk of their time analyzing, producing and executing. They can move from completing tasks to shaping outcomes. They learn to redesign workflows, work across AI systems and operate at a level of thinking that used to belong to roles above them. Among AI users surveyed in the Microsoft research, 58% say they are producing work they could not have done a year ago. Nearly half of all Microsoft 365 Copilot conversations support cognitive work — analysis, problem-solving, evaluation and creative thinking.

The capability is real. The room to act on it is not.

Because the org chart has not moved. Roles still define who owns what. Decision-making authority still follows level. The metrics that determine performance still reflect an older model of the job. Employees are adopting tools and expanding what they can do. But they run directly into the boundaries of their roles, their job descriptions and the authority tied to their level.

They can do more. The organization has not redefined what they are allowed to do, rewarded it or built the conditions to support it.

In a conversation on The Future Of Less Work podcast, Jared Spataro, Chief Marketing Officer for AI at Work at Microsoft, puts it simply: “When the system itself has a governor on the speed that it can go, it doesn’t matter how fast an individual can run.”

The Microsoft 2026 Work Trend Index puts data behind this gap. Only about 9% of AI users at work fall squarely into blocked agency — skilled people whose organizations have not caught up. But the dynamic runs much wider. Roughly half of all AI users sit in what the report calls the “emergent zone,” where individual capability and organizational readiness are both still developing, usually at different speeds, with the person ahead of the system.

The Transformation Paradox: Why AI Adoption And Work Redesign Clash

What makes this a structural problem rather than a motivation problem is the data on what employees are actually experiencing.

65% of AI users at work fear falling behind if they do not use AI to adapt quickly. At the same time, 45% say it feels safer to focus on current goals than to redesign how they work. Only 13% say they are rewarded for reinventing their work with AI, even when results are met.

Taken together, those numbers indicate that people know they need to change. They even want to change. But the systems around them make it safer not to.

Microsoft calls this the “transformation paradox”: the same forces accelerating AI adoption are holding back the redesign that would make it meaningful. Employees are ready to reinvent how they work. But the metrics, incentives and norms around them continue to reward the old way.

This is not just a matter of misaligned incentives or slow adoption. It reflects a deeper tension between improving how work is done and rethinking what work is. As Spataro puts it, “we’re not trying to make the processes that we have just a little bit more efficient… I think the firm of the future is a totally different place.”

That tension shows up clearly in how organizations experience the transformation. Only one in four AI users say their leadership is clearly and consistently aligned on AI transformation. Leaders are also significantly more likely than their teams to say that reinvention feels safe and rewarded. That gap in perception is worth pausing on.

Some organizations have already closed this gap. Microsoft calls them “Frontier Firms,” roughly 16% of organizations where individual capability and organizational readiness are reinforcing each other. Their managers openly model AI use and set quality standards for AI-assisted work. Teams share learnings and mistakes. Experimentation is supported, not penalized. Frontier professionals are twice as likely to say they are rewarded for reinvention regardless of outcome. Culture, manager support and talent practices account for more than twice the AI impact of individual mindset and behavior, 67% versus 32%.

The biggest factor behind whether AI delivers real value at work is not the tool. It is the environment built around the people using it.

How To Fix Blocked Agency: The Organizational Redesign Leaders Must Own

The transformation paradox is a systems problem. Systems do not fix themselves.

Redesigning the system to match the work is the job every leader should be focused on right now. That means the metrics, incentives and expectations need to reward people for changing how they work, not just for maintaining what they delivered before. It means creating the kind of culture that treats AI as a strategic advantage and encourages experimentation. And it means managers who model and incentivize AI use, and talent practices that build skills and create genuine space to apply them.

The potential for employees to make an impact has never been higher. The research makes that clear. What is less clear, inside most organizations, is whether the systems around people are being rebuilt fast enough to capture that potential.

Come back to the manager at the beginning of this article. Promoted for results, now overseeing agents that handle execution, but still measured on outputs rather than on the quality of judgment, direction and ownership they bring to the work. That is a design failure. The role evolved. The system around it did not.

Every leader owns a piece of that system. Every leader should be asking a simple question: what would have to change on my team for the people doing the most capable work to also have the most room to act on it?

That is where the redesign begins.