AI Is Reshaping Work. What Does That Mean for Leaders?

The World Economic Forum found that 63% of employers see skills gaps as a major barrier to transformation, while 46% point to organizational culture and resistance to change. 

Those numbers tell us something important about AI transformation. 

The technology may be moving quickly, but organizations—and people—do not automatically change at the same speed. 

When organizations talk about AI, the conversation usually begins with technology. What platform should we use? Which activities can we automate? How much time can we save? Which functions should adopt AI first? 

These are important questions. But the more I speak with HR professionals and business leaders, the more I believe that the most difficult questions come after we answer them. 

Because AI does not simply change what technology can do. 

It changes what people need to do next. 

When AI Enters the Workplace, Work Itself Begins to Change

Imagine an organization introduces an AI assistant that can help employees summarize reports, analyze information, prepare drafts, and complete repetitive administrative work. 

It works. 

People become faster. 

From a technology perspective, the implementation might already look successful. 

But then an employee asks, “If AI can now do part of what I used to spend several hours doing, what should I be doing with that time instead?” 

A manager begins wondering how performance should be evaluated when employees are producing work together with AI. A fresh graduate wonders how they will develop foundational experience if technology performs many of the junior tasks that traditionally helped people learn. Another employee becomes reluctant to experiment because they fear that demonstrating how much of their job can be automated might eventually make their role unnecessary. 

These are not software problems. 

They are questions about work design, leadership, learning, careers, and trust. 

And I believe HR needs to be much more involved in answering them. 

Productivity Is Only Valuable If We Redesign What Happens Next

One of AI’s strongest promises is productivity. 

If a recruiter can prepare a candidate summary in ten minutes instead of thirty, that creates capacity. If an HR team can analyze large amounts of employee feedback in hours instead of days, that creates capacity. If a manager spends less time preparing repetitive reports, that creates capacity. 

But capacity alone is not transformation. 

Organizations need to ask a second question: 

What are we going to do with the time we just saved? 

If AI frees five hours of work only for those five hours to be filled with more emails, meetings, and administrative activities, the organization has automated a task without necessarily improving the work. 

The opportunity is much bigger. 

Perhaps recruiters can spend more time actually speaking with candidates. HR business partners can spend more time working with managers on workforce challenges. Leaders can spend less time compiling information and more time interpreting it. Employees can devote more attention to problem-solving, customer interaction, creativity, judgment, or relationship-building. 

This is where AI stops being a productivity tool and becomes a work redesign opportunity. 

The question changes from “What can AI do for this employee?” to “What should this employee be able to do more of because AI now exists?” 

Managers May Determine Whether AI Adoption Succeeds

We spend a great deal of time talking about preparing employees for AI. 

We may need to spend just as much time preparing managers. 

Employees look to their immediate leaders for signals about almost every organizational change. Managers influence whether employees feel safe experimenting, whether learning is encouraged, how new expectations are interpreted, and whether productivity improvements feel like opportunities or threats. 

Imagine an organization telling employees to experiment with AI while managers discourage them from changing established processes. Adoption stalls. 

Imagine leaders talking about augmentation while managers immediately interpret every productivity improvement as an opportunity to reduce headcount. Trust disappears. 

Or imagine encouraging experimentation while criticizing employees every time an AI experiment does not work perfectly. Employees quickly learn that the safest option is not to experiment at all. 

Managers therefore need something more than AI tool training. 

They need to understand how to lead when the division of work between humans and technology is constantly changing. 

They need to ask where their teams are spending unnecessary effort, where human judgment remains critical, which workflows should be redesigned, and where AI should assist rather than decide. 

In an AI-enabled organization, these may soon become ordinary management questions. 

AI Fluency Must Include Judgment

There is another change that HR and learning leaders should pay attention to: our definition of what it means to be skilled. 

For many years, expertise often meant being able to produce something yourself. 

AI is beginning to change that. 

Increasingly, expertise will also mean knowing how to direct, evaluate, challenge, verify, and improve what technology produces. 

An employee may use AI to prepare an impressive report. But can they recognize whether the conclusion is wrong? 

A recruitment tool may recommend a candidate. But can the recruiter evaluate whether the evidence actually supports the recommendation? 

An AI chatbot may provide an employee with an answer. But does HR recognize when the situation requires human judgment or empathy? 

This is why I would be careful about reducing AI fluency to prompt engineering. 

Prompting matters. But judgment matters more. 

As AI becomes better at producing convincing answers, people need to become better at questioning them. 

That means the future workforce needs not only technological skills, but also critical thinking, professional expertise, ethical judgment, communication, and the confidence to say, “This output doesn’t look right.” 

We Need a More Credible Conversation About Job Change

We also cannot discuss AI adoption seriously without discussing fear. 

Some employees are worried about what AI means for their careers. 

Leaders may be tempted to respond by saying that people have nothing to worry about. 

I do not think that is always the most credible answer. 

The World Economic Forum reports that by 2030, 59 out of every 100 workers are expected to require training, while employers anticipate both significant job creation and displacement as technologies and other major trends reshape work. 

Some tasks will disappear. Some roles will change. New jobs will emerge. Certain capabilities will become more valuable, while others will become less important. 

So rather than promising that nothing will change, organizations should help people understand how the work may change and how they can remain relevant within it. 

If roles are evolving, employees should know what new capabilities they need. If AI can take over routine tasks, organizations should identify the higher-value work employees can move toward. If internal mobility or reskilling will be required, those pathways should be made visible early. 

Trust is not built by pretending that change will not happen. 

It is built by showing people that the organization has thought seriously about what happens to them when it does. 

HR Cannot Enter the Conversation After the Technology Decision

This is why I believe HR should resist becoming the department that gets called only after the AI platform has been purchased. 

HR should be part of the conversation much earlier. 

What work will AI change? Which tasks should remain human? What skills will employees need? How should roles be redesigned? How should managers lead differently? What should employees be accountable for when AI contributes to their output? How will careers develop when entry-level work changes? 

Technology teams can tell us what AI is capable of doing. 

HR can help the organization decide what those capabilities should mean for people and work. 

That is a much more strategic role. 

The Most Future-Ready Skill May Be the Ability to Keep Learning

There is also a practical reality about AI that organizations need to accept. 

Whatever employees learn today will change. 

Platforms will evolve. Interfaces will change. Models will improve. Capabilities that seem remarkable today may become routine surprisingly quickly. 

That means organizations cannot build AI readiness around mastering one platform. 

They need people who are comfortable learning, experimenting, questioning, adapting, and learning again. 

This shifts the question for L&D. 

Instead of asking, “Have our employees completed AI training?” we should increasingly ask: 

“Are our people becoming more capable of adapting as AI changes?” 

Training completion tells us that learning happened. 

Adaptability tells us whether the organization can continue moving. 

The Real AI Transformation Is Organizational

Good AI adoption still requires strategy, governance, data, measurement, training, experimentation, and continuous evaluation. 

But underneath all of those practices is something more human: the organization’s ability to change how people work without losing their trust, capability, judgment, and sense of purpose along the way. 

This is the opportunity I see for HR. 

We should not fear the technology, nor should we simply celebrate every new capability it creates. 

We should help organizations translate those capabilities into better work. 

Because the most important AI question may eventually stop being: “What can this technology do?” 

The better question may be: “Now that technology can do this, what should our people become capable of doing next?”