The AI Era: Why Managers Must Adapt to the New Atomic Unit of Work

As AI continues to mature, it’s altering what many experts are calling the atomic unit of work—the smallest, most fundamental element that defines how work gets done.

This shift has implications for leaders. We are moving from rigid, role-based organisational models to a more fluid, skill- and task-based ecosystem. For managers, this means rethinking not just what your teams do, but how they do it, what skills they need, and how quickly they must evolve.

What Is the Atomic Unit of Work—and Why Is It Changing?

Traditionally, the atomic unit of work was the job role. Taking a lego block analogy: each block (or person) was assigned a defined function—HR, finance, product, support—with clear boundaries and outputs. AI breaks these boundaries. By making adjacent capabilities accessible to everyone, it empowers individuals to operate beyond their job descriptions. 

An analyst using ChatGPT can now write communications-worthy executive summaries. A product manager using AI design tools can mock up a user interface without needing a designer. A support agent can generate onboarding workflows for a new product.

This isn't just task automation—it’s task expansion. As AI augments individual capabilities, it shifts the atomic unit of work from the role to the skill or even the task. This allows work to become modular, fluid, and rapidly reconfigurable—demanding a whole new way of thinking about team structures and management.

Breaking Job Boundaries: What It Looks Like in Practice

Let’s take a practical example: a marketing team. Traditionally, you might have had a content writer, a data analyst, a designer, and a social media manager. Each person worked within their silo, collaborating at handoff points. In an AI-augmented environment:

• The content writer can use AI tools to generate design ideas or repurpose long-form blogs into social posts.

• The data analyst can use AI to translate performance dashboards into readable narratives for clients.

• The designer can quickly test multiple iterations with the help of generative AI tools and even A/B test creative directly in-platform.

Suddenly, the lines blur. Each team member has visibility into adjacent roles—and, increasingly, the ability to take on tasks previously out of their scope.

This erosion of boundaries doesn’t eliminate jobs. Instead, it reconfigures them. The skills required to be effective are broader and more dynamic. The content writer may now be expected to have some fluency with data and light design. The analyst might need strong communication and storytelling skills.

Role Definitions Under Pressure

This fluidity puts immense pressure on how we define jobs. What does a “designer” mean when AI can handle much of the first-draft ideation? What is the role of a “project manager” when AI tools are able to assign tasks, track deadlines, and even summarise project updates automatically? In many organisations, we’re already seeing a shift:

• Job descriptions are becoming more flexible. Roles are framed around core capabilities with an expectation of adjacent skill application. 

• Performance expectations are evolving. It’s no longer just about deep expertise in one function, but also about your ability to flex, learn, and collaborate across domains.

• Hiring is shifting. Recruiters are beginning to prioritise learning agility and cross-functional thinking over narrow specialisation.

The Rise of Skill Flux

The ability to “flex” into new tasks isn’t just a nice-to-have—it’s essential. This is where the concept of skill flux comes in. Research suggests that the half-life of technical skills is now just 2.5 years—and shrinking. This means that a tool or capability learned today might be outdated before your next promotion cycle. AI accelerates this cycle by commoditizing once-rare skills. Take video editing. What used to require expensive software and technical training can now be done with drag-and-drop AI platforms in minutes. The result? The baseline is rising—and yesterday’s specialist is looking more and more like today’s generalist. 

In this landscape, lifelong learning becomes the price of entry. But traditional models of upskilling—once-a-year workshops, or formal training programs—simply won’t cut it. We need surge skilling: rapid, just-in-time acquisition of new capabilities aligned with immediate business needs. I’ll likely cover this and other practical advice for managers in upcoming posts.