Looking back, I sometimes think the least important part of my first job was the work itself.

I spent most of my days taking notes, scheduling meetings, and sending follow-up emails. Today, AI could probably do 80% of that work better than I did.

At the time, I couldn’t wait to stop doing it. I thought the real work would begin once someone trusted me with bigger decisions. Boy oh boy, was I wrong, that first chapter was oh so needed in my career trajectory (hindsight is 20/20, right?) because……

It got me in the room.

In those rooms, I learned how decisions were made. I learned what leadership looked like. I learned how people hired, how they disagreed (some days were better than others), how they changed their minds, how they handled uncertainty, how they sold an idea without making it sound like they were selling.

No one sat me down and said, “Here is how judgment works.” I learned by watching. And that is the part I’m worried we are about to lose.

Because companies are not just automating tasks. They may be automating the training ground.

This is the problem with calling entry-level work “busywork.” Some of it is busywork. Some of it is boring. Some of it absolutely should be automated. But some of it was also the price of admission to the rooms where taste, judgment, and professional instincts were formed.

The note-taking was not the lesson. The room was the lesson.

I keep thinking about an email I got today from someone who graduated last year from an incredible college. They weren’t asking me for a job. They were asking if they could shadow someone. Just to be in the room. Just to see how things work.

This feels like a bat signal is being flashed. Not about one person’s career path, but about a much bigger shift. What happens when talented young people can no longer get close enough to watch how good work actually happens?

For thousands of years, apprenticeship was how knowledge moved from one generation to the next. You became a blacksmith, artist, physician, writer, founder, or operator by standing near someone who had done it before. You watched what they noticed. You watched what they ignored. You watched when they pushed, when they waited, when they changed their mind.

That kind of learning is hard to put in a course, because the whole point is that the lesson is often happening underneath the words.

Psychologists sometimes call this tacit knowledge: the kind of knowledge people have but cannot always explain. It is the difference between knowing the rules and knowing the room. It is why an experienced doctor hears one strange detail in a patient’s story and suddenly pays attention. It is why a seasoned manager can feel when a meeting is technically aligned but emotionally unresolved.

Expertise doesn’t just change what you know. It changes what you notice.

A beginner sees a meeting. An experienced person sees the real decision happening three minutes before anyone says it out loud. A beginner sees a client question. An experienced person hears the fear underneath it. A beginner sees a decent first draft. An experienced person can feel where the edges are soft.

That is what experience gives you. Not just information, but perception.

And perception is about to become the scarce thing.

Because the future is not going to have a shortage of content, decks, briefs, emails, strategies, or ideas. AI will make sure of that. We are going to have more output than we know what to do with.

The shortage will be people who can tell whether any of it is good.

That is the higher-stakes problem. We are building a world with infinite production and a shrinking pipeline for judgment.

And this is where it gets uncomfortable. A lot of companies will see AI as a way to make leaner teams. Fewer assistants. Fewer coordinators. Fewer analysts. Fewer junior people sitting quietly in meetings doing the thing everyone says is beneath them.

In the short term, that will look efficient.

In the long term, it may create a succession problem.

Who becomes the senior strategist if no one starts as the junior person in the room? Who becomes the great manager if they never watched a great manager handle conflict? Who becomes the person we trust with hard decisions if they never got to see how hard decisions are actually made?

You can read about leadership. You can listen to podcasts about negotiation. You can ask AI to summarize the principles of good management. But at some point, you have to watch someone navigate a tense conversation in real time. You have to see the pause before they answer. You have to feel the mood shift when they ask the right question. You have to notice that the person with the best title is not always the person with the most influence.

Some things have to be lived before they’re understood.

But there is another side to this.

If you already have years of work experience, AI may be one of the biggest advantages you’ve ever been handed, because AI is much more powerful in the hands of someone with taste.

If you ask ChatGPT for a brand brief, it might give you 70% of something useful. But if you have led countless brand briefs, you know what is missing. You know where the logic is thin. You know where the client will get confused. You know what needs more margin, more specificity, more emotional tension, more operational reality.

AI can generate the draft. But your experience tells you whether it is any good.

That is the advantage a lot of millennials and Gen Xers are underestimating. We often talk about ourselves as if we are at a disadvantage because we are not perfectly digitally native. We didn’t grow up with every tool already in our hands. We may not instinctively understand every platform shift or AI product the second it launches.

But we have something else: years of IRL pattern recognition.

We know how work actually gets done. We know how people actually make decisions. We know that a great idea still has to survive a budget meeting. We know that a strategy deck is only as good as the person who can get people aligned around it.

That matters more than we think.

AI can give you speed, but it cannot give you taste you never developed. It can give you options, but it cannot always tell you which option fits the moment. It can summarize the meeting, but it cannot always sense the thing no one wanted to say out loud.

That is judgment. And judgment is becoming more valuable, not less.

So yes, we have an apprentice problem. Young people need access to rooms, not just tools. They need to see how judgment is practiced in real time.

But we also have a responsibility problem. Those of us who learned by watching may now have to become much more intentional about letting others watch us.

Because if we don’t rebuild apprenticeship in the age of AI, we may end up with faster companies, cleaner decks, fuller calendars, and fewer people who know what any of it is for.

Maybe entry-level jobs were never really about maximizing productivity. Maybe they were how we produced the people we would one day trust to make difficult decisions.

And maybe the next advantage won’t belong to the person who knows the most tools.

Maybe it will belong to the person who knows what good looks like.