Recompose the Role Around the Person, Not Just the Work

Some of the most useful thinking on AI right now is not about the technology at all. It is about people. A recent piece from PwC, Your AI Strategy Has a People-Shaped Hole in It, by Shebani Patel, who leads Workforce Solutions at PwC US, and Dan Priest, its Chief AI Officer, opens with a question they now ask CEOs. You have told me what your AI can do. Now tell me specifically what your people do differently because of it.

Most, they report, cannot answer with any precision. They can describe the agents deployed and the pilots launched. They cannot describe how anyone's Tuesday has changed.

It is a brilliant diagnostic question, and the framework behind it is just as strong. PwC organizes AI transformation around three dimensions: the work, the workforce and the worker. Most companies, the authors argue, invest heavily in the first and barely address the other two. That is where the value of AI either comes to life or quietly dies.

I agree completely. And I want to build on it.

The Evidence That Judgment Is Rising in Value

PwC's own research supports the case. According to its 2026 Global AI Jobs Barometer, jobs that AI professionalizes, raising the premium on human expertise and judgment, are growing twice as fast as jobs AI simplifies, with 42% faster wage growth since 2021.

In other words, the market is already rewarding what makes people distinct. The question is whether organizations will design for it.

This is the same pattern I wrote about in AI Is Producing Motion. The Return Depends on Meaning. The machine multiplies output. The return depends on the human system around it.

Outside In or Inside Out

PwC's answer to the worker dimension is specific and practical. Recompose each role around new work. Define the skills precisely. Align incentives so people are rewarded for judgment rather than volume. Earn adoption instead of mandating it.

All of that is right. But notice how most of it is designed from the outside in. The organization decides what the new role is, what skills it requires and how it will be measured. Then it hopes people will opt in.

The authors point to the deeper truth themselves. People will embrace AI, they write, when they see recomposed roles they can see themselves in, and when they have the agency to shape how AI shows up in their work. That is the turn that matters most.

In my work on Leadership in the Age of Personalization, the central argument is that it is no longer about institutions defining the individual. It is about the individual defining the institution. A recomposed role that is designed without knowing the person in it is still standardization, just a more modern version of it. Standards still matter. They simply work best when they are built with the people who will live inside them.

Consider PwC's underwriter, whose week shifts from data gathering to edge cases, portfolio patterns and broker relationships. Two underwriters in that same role may bring very different strengths. One sees patterns others miss. Another builds trust with brokers faster than anyone in the building. Designing the role around the work alone treats them the same. Designing it around the person lets each one contribute what they uniquely bring.

That is why the first question I ask any leader is not what they do. It is what they solve for.

Results to Methods

PwC makes one of the sharpest points I have read on this subject. If you redesign a role around judgment but still measure throughput, you have "designed the new work but incentivized the old work."

This is exactly the fifth shift in my methodology, from results to methods, and the question beneath it: How do you let them do it? Organizations that reward only volume will keep getting volume, no matter how sophisticated their AI becomes. The leaders who win will measure how good decisions get made, not just how many tasks get closed.

The fourth shift, from mission to contribution, asks a related question: What do you let them do? A recomposed role is ultimately a contribution decision. As I explored in You Are What the AI Efficiency Calculation Leaves Out, the most valuable things people contribute rarely show up in the efficiency math.

Three Questions to Add to the Diagnostic

PwC closes with six excellent questions for leadership teams. I would add three more.

Do we know what each person in a recomposed role solves for, or only what the role requires?

Who designed the new role: the organization alone, or the person together with the organization?

Are we measuring volume, or the quality of judgment we say we value?

The information required to answer them almost always exists inside the building. It lives with the people doing the work. The organizations that ask them will not just fill the people-shaped hole in their AI strategy. They will build that strategy around the people who make it work.

Want to explore these ideas further? Learn more about my work on leadership, identity and conviction at www.theglennllopis.com.