How Expertise Develops in an AI-Enhanced World

In my previous post, I explored why human judgment still matters in AI-assisted work.

That led me to another question.

If judgment matters, how do people develop it?

Most of us are not taught judgment directly. We develop it slowly through experience.

We do the work. We see different situations. We make mistakes. We receive feedback. We notice patterns. Over time, we start to understand what matters, what can go wrong, and what to do next.

That process is not always efficient.

But it works.

As AI becomes more involved in day-to-day work, I have started wondering whether that pathway is changing.

The Traditional Path to Expertise

In many roles, people begin by doing simpler or more routine tasks.

Those tasks may not seem exciting, but they are important.

They give people exposure.

A junior employee might start by preparing drafts, reviewing notes, organizing information, checking details, or supporting more experienced team members.

At first, they may only see the surface of the work.

But over time, they start noticing more.

They learn what good looks like.

They see what mistakes happen often.

They learn when something is unusual.

They begin to recognize patterns.

Eventually, they are not just completing tasks anymore. They are developing judgment.

This is one way expertise grows.

What AI Changes

AI is very good at helping with routine work.

It can summarize notes, organize information, draft content, generate options, and identify patterns.

That can be very useful.

But there is a learning question hidden inside this shift.

If AI starts doing some of the routine work that people used to learn from, where will people get the exposure they need to develop judgment?

This is especially important for people who are newer to a role.

They may be expected to make decisions earlier because AI can help with basic execution. But the experiences that used to help them build judgment may no longer happen in the same way.

That creates a new challenge.

The entry point into effective work may be moving higher, while the traditional path to gaining experience becomes weaker.

A Coaching Example

This idea became clearer to me through the same coaching workflow project I mentioned in my earlier posts.

At first, the project was about using AI to support coaching preparation.

Then it became about workflow.

Then it became about judgment.

But there was another layer.

The AI-assisted workflow included information from the coaches’ playbook, which contained coaching standards and best practices.

One experienced coach made a comment that stayed with me:

“The GPT has read our coaches playbook, so if we are doing something that isn’t living up to that standard, it can call us out on that.”

What interested me was not only that AI could remind coaches of the standard.

It was that some of the expertise that usually lives in practice, memory, and experience was becoming more visible.

Even experienced people forget things.

They get busy.

They focus on one part of the work and miss another.

They develop habits.

If AI can surface useful reminders, questions, and patterns at the right moment, it may help people apply what they already know more consistently.

For newer people, it may also help them see what experienced people are paying attention to.

A Different Way to Think About Learning

Traditionally, expertise often develops through accidental exposure.

You encounter a difficult situation because it happens in real work.

You learn from it because you had to deal with it.

But not every important situation happens often.

Some edge cases may only show up once in a while.

Some mistakes are too costly to let people learn from directly.

Some judgment takes years to build because people need time to see enough variation.

This makes me wonder whether AI can help create more intentional exposure.

Not by replacing real experience.

But by helping people practice with more examples, more scenarios, and more feedback before they face those situations in real work.

Maybe expertise development does not have to depend only on time and chance.

Maybe some parts of experience can be designed more intentionally.

What I Am Still Thinking About

I do not think AI removes the need for real experience.

A simulation is still a simulation.

A reminder is not the same as lived judgment.

A suggested question is not the same as knowing how to handle the answer.

But I do think AI may change how people begin developing expertise.

It may help make expert thinking more visible.

It may give newer employees more chances to practice judgment before the stakes are high.

It may help people encounter more variation in less time.

That does not mean expertise becomes easy.

It means the path may look different.

Final Thoughts

AI is changing how work gets done.

But I think it may also change how people become good at their work.

If routine tasks become more automated, organizations may need to think more carefully about where experience comes from.

If people are expected to use judgment earlier, organizations may need to create better ways for them to practice judgment earlier.

I do not have a final answer yet.

But I am becoming more convinced that AI adoption is not only about improving work.

It is also about rethinking how people learn to do the work well.

And that leads to another question.

If we create faster pathways to experience, what might we lose along the way?