AI isn't going to take a job worth doing. It's going to expose the jobs that don't require critical thinking and communication — and hand the upside to the people who can do both.
That part isn't new. That's what every technology revolution has done.
What is new is how fast it's happening, and how badly most leaders are misreading what their job actually is in the middle of it.
I've been thinking about this from two directions.
One is professional. A marketing leader told me last week, three weeks into his AI rollout: "The people who are going to struggle are the ones who've been hiding behind the process." He wasn't being harsh. He was being precise. Every team has people whose contribution is real but whose visibility comes from managing the machinery of work — coordinating, formatting, shepherding, moving things from one inbox to another. When the machinery gets automated, what's left is the thinking. And a lot of smart, hardworking people have never had to prove they're good at that part. Not because they can't. Because no one ever asked them to.
The other is personal. I've been spending time this spring at NESCAC colleges, particularly Hamilton. The focus there is exactly what you'd hope — writing, communication, critical thinking, Socratic method. These students are being prepared for the one thing we know for sure about the next decade: that the ability to think clearly and communicate precisely will be irreplaceable.
Hold those two things next to each other. A workforce promoted for executing process. A generation educated for asking the right question. Now drop AI into the middle.
Joelle kaufman
This is the moment. And it's not the first time we've been in it.
The Industrial Revolution Wasn't Won by the Steam Engine
The standard story is that steam, coal, and iron made the modern world. They didn't. They made a possibility. Whether a country captured the upside depended almost entirely on what its leaders did about the people.
Britain industrialized first and dominated the 19th century. Prussia industrialized second and built the dominant economy of the 20th. The difference was not technology. The difference was that Prussia, starting in the early 1800s, built the world's first compulsory public education system — explicitly designed to produce literate, numerate workers who could follow written instructions, do basic arithmetic, and operate inside a hierarchy. Reading. Writing. Math. Discipline.
That's not a romantic origin story. It was a deliberate institutional bet on the skills the new economy would reward.
Britain assumed the workforce would adapt on its own. Prussia decided to build the workforce on purpose. By 1900, Germany had passed Britain in steel production, chemical manufacturing, and engineering exports.
The lesson the leaders of that era kept missing: the hard part wasn't the technology. The hard part was the people, and whether someone in charge had decided in advance to build them for what was coming.
The Digital Revolution Wasn't Won by the Microchip
Skip ahead 150 years. After World War II, the United States made two leadership bets that shaped the next half-century.
The first was the GI Bill — sixteen million veterans handed access to college, vocational training, and home ownership. The second was the explosive expansion of community colleges through the 1960s and 70s, which took higher education from an elite credential to a mass-market skill-building system.
These weren't technology investments. They were people investments, made decades before the personal computer existed and a full generation before anyone outside a research lab had heard the word "internet."
When the digital revolution hit in the 1980s and 90s, the United States had something most other developed countries didn't: a workforce that had been steadily upskilled across two generations. People who could learn new software. Switch industries. Read a manual. Take a continuing education class at night.
Other countries had the same chips. They didn't have the same readiness.
Meanwhile, the companies that dominated the digital era weren't the ones with the best technology. They were the ones whose leaders treated workforce development as a strategic capability, not an HR line item. Intel's continuous learning culture. Microsoft's certification ecosystems. Cisco's massive investment in technical training, including for people who didn't yet work for them. These leaders understood something the slower companies didn't: in a transition, the organization's ceiling is the skill ceiling of the people inside it.
The companies that lost the digital era didn't lose because they couldn't buy the technology. They lost because their people weren't ready to use it well, and their leaders hadn't done the work to change that.
The AI Revolution Will Not Be Won by AI
Which brings us to now.
Right now, most leaders are treating AI as a technology problem. They're buying licenses. They're standing up centers of excellence. They're naming AI champions. They're rolling out pilots.
None of that is wrong. All of that is insufficient.
Because the actual question — the one history keeps asking and most leaders keep failing — is not what tools are you deploying? It's what people are you building?
The Prussian state didn't ask "how do we get more steam engines?" It asked "what skills will every citizen need to thrive in the world we can see coming?" — and built an institution to deliver those skills at scale.
The post-war U.S. didn't ask "how do we put a computer on every desk?" It asked "how do we make sure tens of millions of people can keep learning for the rest of their lives?" — and built the institutions that made the digital era possible thirty years later.
The leaders who win the AI era won't be the ones who deployed first. They'll be the ones who decided, before the rollout, what their people needed to become — and built the container to make it possible.
What That Looks Like in a Real Company
Back to that marketing leader.
His AI rollout is going well. Not because of the technology. Because eighteen months before the rollout, he made his team go through a talent assessment. Everyone was required to write their own development plan — not one handed down from HR, one they owned. He had honest conversations about what skills would matter in eighteen months: critical thinking, business acumen, the ability to connect their work to the outcomes the business actually cared about.
He didn't know AI was coming the way it came. But he had already decided, as a leader, that his job was to build people who could think — not just people who could execute.
When the technology arrived, he had a workforce that was already in motion. The tools amplified what was there. They didn't expose a gap.
Most of his peers didn't do that work. They're now discovering, in real time, that their teams were promoted for the wrong things. That their high performers were often just their best executors. That the cultural muscle they'd need to navigate this — honest skill conversations, individual ownership of growth, comfort with reinvention — was never built.
The technology surfaces what's there. The culture has no way to hold it.
This is the Talent Trap in its most modern form. Leaders built teams around task completion, then wondered why the team couldn't think strategically. The tasks were never the job. They were the camouflage.
The Reframe
If you're a leader right now, the question is not what's your AI strategy?
The question is: what is your people-readiness strategy, and would it survive a hard look?
Specifically:
- Have you named, out loud and more than once, the skills that will determine your team's relevance in eighteen months? Not vague ones. Specific ones — judgment, business acumen, the ability to ask the right question before anyone else has framed it.
- Have you made development something your people own, or is it still something that happens to them? Required individual development plans, written by the individual, are not a perk. They're the institutional move that determines whether the next transition exposes your team or rewards it.
- Have you separated the work from the worker? Have you actually looked at what your team spends time on and asked which tasks require genuine judgment and which are process-dependent? The answer will tell you more about your team's real skill distribution than any performance review ever has.
- Have you been honest about which of your top performers are great executors and which are great thinkers? Both have value. Only one set survives the transition without help.
The leaders who get this right aren't the ones moving fastest. They're the ones who started building the foundation before they needed it — the way Prussia built schools before the steam engine arrived in force, the way the U.S. built community colleges before the chip changed the economy.
The Test Every Era Sends
Every technology revolution sends the same test to leaders: will you treat this as a tools problem or a people problem?
The leaders who answer "tools" lose. Not immediately. But within a generation, every time.
The leaders who answer "people" — and then do the unglamorous, multi-year work of building a workforce that can think, learn, and adapt — are the ones whose institutions, companies, and countries dominate what comes next.
The kids walking out of Hamilton this May are going to be fine. The question is whether the people already on your team will be — and whether you've given them any reason to develop the skills that actually matter now.
What would a hard look at your team's work reveal about the skills you've been rewarding?
And what's stopping you, as a leader, from changing the answer?
Read more
The Talent Trap: Why AI Rollouts Are Really About Clarity