Every major technological wave — from digital to mobile to cloud — followed the same IT playbook: select a platform, run pilots, write policies, train employees. That approach worked when technology simply digitized what already existed. But AI is different. Instead of sitting behind the work, it reshapes who does it, how decisions are made, and even what ‘work’ means. It’s not just a technology upgrade — it’s an organizational design challenge.
The limits of the old playbook
When employees work with AI, ownership blurs. Who is responsible for an outcome — the person, the model, or both? Who retrains the system when it learns the wrong lesson? Who decides when to trust a recommendation and when to challenge it? These are not IT questions; they’re questions for the people closest to the work — the ones who understand how value is actually created.
This is why the biggest risk isn’t technical. It’s human. During a recent conversation on The Future of Less Work podcast, Mary Alice Vuicic, Chief People Officer at Thomson Reuters, explained that AI transformations succeed or fail not on the technology but on how well people adapt. At Thomson Reuters, the CIO and CHRO co-sponsor the company’s AI transformation framework — a model for shared ownership between technology and talent. “This is about two equal parts technology and change management,” she said.
What redesign really looks like
In past transformations, people were users of tools. In this one, they are partners in a system that learns alongside them. Recruiting offers a clear window into what this means. For years, recruiters spent most of their time managing transactions — screening résumés, scheduling interviews, collecting feedback, and coordinating across systems. It was necessary work, but largely administrative, and it kept recruiters focused on matching candidates to job descriptions rather than matching capabilities to organizational needs.
AI now performs much of that coordination instantly — summarizing résumés, scheduling interviews, and synthesizing feedback. Freed from administrative tasks, recruiters can focus on what humans do best: understanding potential, not just experience. Their role expands beyond filling jobs to orchestrating how skills flow through the organization — connecting opportunity with capability and ensuring the company has the talent it needs to evolve. They may even stop being “recruiters” in the traditional sense and become something broader: talent architects who design the systems and experiences that help the organization compete and grow.
This evolution doesn’t diminish expertise; it expands it. The recruiter becomes the designer of a process where technology amplifies human judgment rather than replaces it.
Vuicic described this as removing the “no-joy” work — the repetitive tasks that drain energy but don’t build capability. In one acquisition, that idea came to life “through the initiative of an individual being curious about what it [AI] could do for us,” Vuicic recalled, “they eliminated 95% of that administrative work and we got a higher quality output… in a fraction of the time. That enabled people to focus on the higher value, more strategic aspects of the acquisition: How do we make sure that we’re retaining the talent and the critical competitive advantages that this organization brings?”
The AI Work Redesign Loop
This kind of transformation can’t be delegated to IT. It must be led by the people closest to the work — the ones who understand the processes, decisions, and skills that create value. A useful way to approach this is through an AI Work Redesign Loop — a framework for rethinking how humans and systems share work.
The first step is to start with outcomes, not activities. Ask what success really means in your area — not in terms of speed or volume, but in terms of quality, insight, and impact. A sales team might define success as deeper customer relationships, not just higher call volumes. A healthcare organization might define it as better recovery rates, not just shorter visits. The right outcomes — not efficiency alone — should guide where AI will make the biggest impact.
Next, deconstruct the work. Look at every task and ask: what can be automated, what should be augmented, and what must remain distinctly human? Seeing work through this lens reveals how AI changes not just the tools you use, but the nature of your contribution.
Then, design the collaboration. Decide how humans and AI will interact day to day. What do you still need to verify or interpret yourself? Where do you need context that a system can’t see? What data or guardrails do you need to make those calls confidently? This is where new skills emerge — not just technical fluency, but the ability to question, guide, and communicate how AI fits into decisions.
Finally, rethink how success is measured. Old metrics tend to reward efficiency: how many calls, how fast the turnaround, how low the cost. In an AI-enabled environment, better metrics capture quality, trust, adaptability, and learning. Ultimately, your most valuable measure may not be how much you automate, but how well people and systems learn from each other.
When you finish this kind of reflection, you end up with something more powerful than a workflow map. You see a work design — a living system of work that clarifies how humans and AI contribute together, what skills people need next, and how the work should keep evolving.
Mapping workflows makes what exists more efficient.
Mapping work redesign makes what’s possible more real.
Why AI Demands a Redesign of Work
In past transformations, we gave people new tools. In this one, we’re giving them new teammates. And if you don’t onboard those teammates properly — or teach people how to collaborate with them — you’ll never get the value you expect.
AI distributes intelligence — compressing hours into moments and freeing people for curiosity, creativity, and connection. But that opportunity also carries a risk: when organizations don’t redesign how work and decisions flow, the technology advances while people remain stuck in old systems.
That’s why every organization needs an operating model for AI — one that aligns people, processes, and intelligent systems in a shared flow of work. The organizations that thrive won’t be the ones that automate the most. They’ll be the ones that reimagine how people and technology create value together — where AI handles scale and humans handle meaning.