Every week, Built. Not Born. examines an assumption about work or leadership that deserves a harder look. We use current research, practical experience, and timeless principles to help you build better teams, stronger organizations, and, most importantly, a life by design.

There's an assumption gaining ground in organizations everywhere: if artificial intelligence helps employees complete work faster, those employees should be able to accomplish more in the same amount of time.

That sounds reasonable until you consider the work required to make AI useful. Someone still has to learn the tools, evaluate their output, correct their mistakes, and determine when the technology is creating more problems than it solves.

I've spent years watching organizations introduce new expectations without reconsidering what their people need to meet them. AI is making that leadership problem harder to ignore.

Here's the question I want you to consider this week:

The promise of artificial intelligence is difficult to ignore. Work that once took hours can sometimes be completed in minutes. Software developers can generate code from a prompt. Sales teams can prepare account research faster. Managers can summarize information, develop plans, and automate administrative tasks that once consumed significant portions of their day.

There's real value in those capabilities, and I don't believe leaders should dismiss them. But I am concerned about the assumptions organizations are making as they introduce AI into everyday work.

A tool that produces an answer in thirty seconds hasn't necessarily completed thirty minutes of work. Someone may still need to determine whether the answer is correct, whether it addresses the actual problem, and whether using it creates consequences nobody anticipated.

Yet organizations are already making decisions about productivity based on the speed of the technology rather than the total work required to produce a reliable result.

Research is beginning to expose the gap between AI's promise and what employees experience.

In Stack Overflow's 2025 Developer Survey, 66% of developers identified AI-generated solutions that were nearly correct as a major frustration. Another 45% reported that debugging AI-generated code was more time-consuming.

Consider what that means for an engineering team. AI can produce hundreds of lines of code in seconds, but an experienced developer may then spend hours determining why it doesn't work as expected. The organization sees faster code generation. The engineer experiences additional review, correction, and rework.

A separate 2025 study from METR, a nonprofit AI research organization, produced an even more surprising finding. Researchers observed 16 experienced developers completing 246 tasks in software projects they knew well. The developers expected AI to make them faster, but their actual completion times increased by 19% when AI tools were available.

That finding doesn't mean AI makes every developer slower. The research examined particular developers, projects, and tools at a specific point in time. But it challenges an assumption that deserves scrutiny: giving someone a faster tool automatically makes their work faster.

The speed of the tool is not the same thing as the productivity of the person using it.

And that distinction matters well beyond software engineering.

Imagine asking a sales manager to adopt an AI system that promises to reduce the time spent preparing forecasts. The system generates the report in minutes, but the manager still has to check the assumptions, reconcile inconsistencies, and determine whether the conclusions reflect what's actually happening in the business.

If those additional responsibilities aren't considered part of the work, the organization may conclude that the manager now has more time available. The manager, meanwhile, may be spending just as much time producing a forecast they can trust.

The same problem appears in engineering, marketing, finance, and other functions adopting AI. Producing something is becoming easier. Determining whether that output is useful, accurate, and appropriate still requires human judgment.

I think leaders need to ask a more demanding question than whether employees are using AI. They need to understand what work AI has eliminated, what work it has changed, and what new responsibilities it has created.

Otherwise, we risk rewarding the appearance of greater productivity while overlooking the additional work required to deliver quality.

There's another part of this discussion that concerns me even more.

The Conference Board published research in July 2026 showing that while more than half of surveyed workers regularly used AI, only 33% had participated in employer-provided AI training during the previous six months. Just 48% agreed that their organization provided sufficient time during working hours to develop AI skills.

Think about the contradiction. Organizations are introducing technology that changes how employees perform their jobs, yet fewer than half of the workers surveyed believe they're being given enough time to learn it.

Learning a new technology isn't simply a matter of watching a tutorial. Employees have to understand where the tool works, where it fails, and how to incorporate it into responsibilities for which they're still accountable. As the technology changes, much of that learning has to happen again.

I believe individuals have a responsibility to keep developing their skills. That responsibility doesn't disappear because an employer provides training, and nobody should expect an organization to manage every aspect of their professional growth.

But when a company makes a new capability necessary to perform a job, the company assumes responsibility for creating reasonable conditions in which people can develop it.

You cannot demand that people change how they work while pretending the time required to learn doesn't count as work.

If employees are expected to meet yesterday's deadlines, deliver tomorrow's productivity gains, and acquire entirely new capabilities without any adjustment to their workload, leadership has created a problem that individual effort alone cannot solve.

This is where the conversation becomes uncomfortable because supporting employees through change requires leaders to make choices.

If your team needs time to learn new technology, something else may need to move. A deadline may need to change. A productivity target may need to be reconsidered. A manager may have to explain why output temporarily declines while employees develop capabilities the organization expects to benefit from later.

Those aren't technology decisions. They're leadership decisions.

It's easy to announce that AI will improve efficiency. It's harder to acknowledge that reaching those improvements may require an investment of time before the benefits appear.

The organizations that manage this well won't necessarily be the ones that adopt every new tool first. They'll be the ones whose leaders understand what the technology can do, what their employees still have to do, and what it will take to close the distance between the two.

That requires more than enthusiasm for AI. It requires a willingness to make expectations match reality.

If you're responsible for a team adopting AI, take a closer look at one workflow this week. Don't simply ask whether the technology is saving time. Ask employees what they're doing differently, how much time they're spending checking and correcting output, and what they've had to learn to use the technology effectively.

Then examine whether your performance expectations reflect that work. If employees are spending meaningful time developing new capabilities, have you accounted for it in their workload? Are you giving them room to experiment, make mistakes, and learn without treating every hour away from immediate production as lost productivity?

If you're an individual contributor or preparing to become a manager, start paying attention to the difference between generating an output and delivering a result. Learn to explain where AI genuinely helps, where human expertise remains necessary, and what it takes to produce work you can stand behind.

The question isn't whether your organization should embrace AI. It's whether your organization understands what responsible adoption actually requires.

On On the Shoulders of Giants, I want to take this discussion further by examining what happens when organizations expect employees to master rapidly changing technology without making room for the learning.

This is becoming a real leadership challenge across industries. The pressure to adopt AI is increasing, but the people responsible for making it work still need time, experience, and support to develop sound judgment about when to trust it.

The podcast gives us an opportunity to explore the difficult choices leaders face when the technology moves faster than their organizations can reasonably adapt.

Whether you prefer to watch or listen, we'd love to have you join the conversation.

🎙️ Listen on Apple Podcasts, Spotify, or wherever you subscribe to podcasts.

📺 Watch on YouTube @HighAchieverLeadership.

If someone forwarded Built. Not Born. to you, we'd love to have you join us. Every week, we challenge an assumption about leadership or work and explore principles that help people become better leaders by design.

If this week's issue made you reconsider how your organization is introducing AI, share it with someone responsible for leading people through change.

For years I’ve said: Poor leaders hurt people. Great leaders change lives. .

Technology will continue changing how we work. Some of those changes will create extraordinary opportunities, while others will force people to rethink skills and habits they've spent years developing.

Great leaders don't pretend those transitions are effortless. They recognize that asking people to work differently also means taking responsibility for helping them become capable of doing so.

If AI is going to change the future of work, leadership has to change how it prepares people for that future.

Until next week, keep building.

- Jeff Wells