October 28, 2026 • 5 Min Read

Manucci has experienced that possibility firsthand. With AI’s help, he can now code and build dashboards, exploring fields that once lay beyond his professional reach. He describes it as opening new windows in his profession and his life.

Now imagine giving people across an organization that room to explore things that were once beyond their reach. What could they contribute beyond the work they already know how to do? And would their organizations recognize and make room for those contributions?

That possibility anchored a recent RGP panel discussion, Humans Still Required: The New Rules of Work in the Age of AI. Six leaders in AI, talent and business transformation considered what people and AI can create together, and how employers need to be involved for that potential to take shape. Their discussion touched on questions about expertise, learning and the opportunities people have to grow through their work.

“This shift is deeper than just how we work,” Manucci said. “It’s about who we are at work.”

Opening New Windows in People’s Work

Manucci believes new capabilities emerge through exchanges between people and AI. As people discover they can do something previously beyond their reach, their role can change. They may see a different problem worth solving, propose an idea they once would have left to someone else or take on work they had never considered. Organizations, in turn, need to give those contributions a place in how work gets done.

Manucci calls the resulting capability hybrid competitiveness. People bring expertise, intuition, intention and connection. AI brings analysis, data management and production capacity. The value develops through how those capabilities come together to produce useful, distinctive responses to the needs of customers, colleagues and others.

The opportunity is expansive for business leaders and for organizations. Realizing it requires being specific and intentional in how people and AI will work together.

Deciding Where the Capacity Goes

For Amadally Hosseinbukus, RGP’s managing director of data analytics and AI, the first question is how leaders define productivity.

AI is helping people get more done. In RGP’s global study of 500 professionals, which informed the discussion, 80% said the time it saves fills with more work. Converting those gains into a business result requires a deliberate decision about where the capacity should go.

“Doing the same task faster is not always the same thing as creating enterprise value,” Hosseinbukus said.

Cost savings are one possibility. Innovation, product development, revenue growth and better risk management may offer greater returns. Leaders need to make those choices explicitly and measure the outcomes they want to achieve.

He also drew a distinction between inserting AI into a suboptimal process and rethinking the end-to-end workflow. The former may produce incremental gains. The larger opportunity comes from stepping back, fixing the data and reimagining how the work should happen.

Pooneh Mohazzabi, an AI, transformation and change management expert, has seen organizations struggle when the technology arrives before a clear business case. Employees are asked to find savings, but no one owns the benefit or decides how to use it.

Someone may save 20 minutes a day. How does that become capacity for a new product, a better service or the future organization leaders have in mind?

Mohazzabi pointed to the value of process reengineering before and alongside AI adoption. Understanding the whole process gives businesses a stronger starting point for deciding what the technology can enable.

Jessica Block, RGP’s chief AI officer and interim CFO, recommended making the intended results as empirical as possible. What would improve if the organization had more capacity? Revenue? Margin? The speed of the financial close?

Those conversations can begin before leaders know exactly how each person’s work will change. They give employees and managers a shared basis for discussing new ideas and greater autonomy. They also help a business prepare to use the capacity people create.

Manucci captured the underlying choice: “Time saved is not value by itself. It becomes value when we decide what to use it for.”

Learning by Taking Things Apart

AI changes the nature of how people learn. Manucci described a transition from cumulative value to generative value. Cumulative value is the knowledge people build through years of study and work. Generative value comes from engaging with available knowledge to explore, learn and develop solutions to new demands.

He believes that distinction has implications for how companies recognize contribution. An employee’s experience matters, but so does what they can create through the dialogue with AI.

“If productivity metrics focus only on volume and speed, AI naturally wins,” Manucci said.

Block’s response is to embrace learning in that environment.

“Let’s accept that often AI can give you the answer,” she said. “So it is incumbent upon you to understand why.”

Her approach is to become a reverse engineer. Start with something that works and take it apart, bit by bit, to understand why it works. That is how she learned to code and how she developed intuition she could use later.

She sees familiar parallels in professional life. A consultant learning a client’s business, a lawyer taking on a case or a CFO acquiring a company often starts with the whole and picks it apart. AI offers another way to learn through that kind of investigation.

The teaching needs to reflect it. Employees should learn to document assumptions, explain the stages of work delegated to AI and understand the inputs behind a result. A forecast is useful only if the person using it can make sense of the circumstances under which it might be right.

Block described an internal leadership training exercise that helped supervisors practice those conversations. Participants reviewed simulated financial commentary containing problems such as shortcut assumptions or a previous month’s cash balance. They had to find the errors and probe how the work had been produced.

The expectation was familiar: show your work. Supervisors could hold someone accountable for the process while making AI use an ordinary, open part of the conversation.

For Block, this is an exciting prospect. People entering a field now have a tool that can help them learn faster and spare them some of the drudgery earlier generations endured.

“I don’t mourn the loss of bad work,” she said.

Roxanna Flores, head of people advisory for North America at The Trade Desk, brought another consideration to that optimism: the common sense and intuition people acquire by working through a problem, making mistakes, getting feedback and trying again.

As entry-level work changes, employers will need to build those experiences into development in other ways.

Flores described witnessing the growth that comes from responsibility with support. Someone brings an employee into a difficult conversation, explains the thinking behind a decision or gives them a stretch assignment and stays close enough to coach.

In HR, AI might help prepare a workforce analysis. An emerging professional still needs to understand the underlying data, consider what is missing and work with a leader to decide what action makes sense.

Rotations, project work and intentional career moves can broaden that exposure. Managers also need time and accountability for developing people.

As Flores observed, “Coaching cannot be something organizations simply hope happens between meetings.”

The opportunity is to give people meaningful responsibility earlier, with guidance that helps them build the judgment they will eventually be expected to use.

Bringing the People Closest to the Work into the Design

Pat Reyes, RGP’s global change management leader, emphasized the knowledge employees already hold: the exceptions, relationships and decisions that require judgment, along with what it takes to get the work done.

Those lived experiences belong in the design of new roles and workflows.

Too often, she said, organizations spend considerable time designing a future state and bring employees into the conversation when it is time to implement. With AI, people need to participate from the beginning.

Reyes described a client that introduced a workforce tool and then needed help preparing employees for new responsibilities. Learning to use the technology was only part of the assignment. End users also needed a new set of skills for the tasks they would now own.

That is why involvement needs to continue through experimentation and feedback. Ambassador and citizen networks can give employees ways to challenge assumptions, refine the approach and help colleagues navigate change. Showing people what changed because of their input builds ownership and trust.

For Reyes, change management is also becoming less of a journey between a fixed current state and future state. There is always a next thing to understand and prepare for.

Mohazzabi brought that discussion back to the operating model. Leadership conversations about tools, technology and processes need to explain the people side of change: which tasks humans will handle, what agents will do, how they will interact and who will be accountable.

She urged leaders to communicate those choices early and honestly, including potential effects on jobs. Employees need a safe environment in which to experiment and ask questions, along with a clear account of where the organization is going.

Reyes summed up the ambition: help shape the transformation.

“Adoption matters,” she added, “but adoption alone does not establish success.”

Making Judgment Visible

The partnership with employees also takes shape in everyday conversations about AI use.

Flores described the questions employees may be asking themselves. Will a manager think they are resourceful? Will a colleague think they took a shortcut? If an assignment takes less time, will someone question why they are needed?

People pay attention to what leaders do and what happens to colleagues who speak openly. Encouraging experimentation sends one message. Responding to an AI-assisted assignment by asking why it took so long sends another.

Leaders can make those conversations easier by showing their own thinking: where AI helped, what it missed and where they applied judgment and experience.

Hosseinbukus added that the scrutiny should fit the task. An email summary and a credit decision call for different levels of explainability and traceability. Hiring recommendations require an understanding of where the data came from and how it informed the result.

That oversight continues as the technology changes. “AI is not a fly-and-forget technology,” he said. Organizations need to monitor performance, bias and unexpected behavior over time.

Together, these examples make human contribution easier to see. It is there in the question someone knows to ask, the assumption they recognize as weak, the context they bring to an analysis and the idea they can now explore.

Manucci’s image of opening new windows gives that contribution room to grow. People can bring their experience into unfamiliar territory, learn through dialogue and discover something they could not have produced before. Leaders can help by giving them the time, support and responsibility to see where it leads.