Ask a twenty-four-year-old which jobs AI is going to take, and you will probably get an answer you dismiss.
You should not dismiss it. They are getting it right, and payroll data now proves it.
In June, the SimplyWise Trades and Technology Index surveyed 1,140 Americans aged 18 to 29 and asked them to rank which occupations AI would disrupt over the next decade. They put software developers, accountants and graphic designers in the highest-risk tier. They put HVAC technicians and plumbers among the safest. Forty-six per cent said the skilled trades feel safer than office work in an AI future, and among those currently enrolled in college it rose to 52%.
It is easy to file that under youthful anxiety. A poll of people who have not yet had a career telling you what careers are worth having.
Then look at what the payroll records say.
The data they did not have access to
In August, the Stanford University‘s Digital Economy Lab published a revised study called “Canaries in the Coal Mine?”, by Erik Brynjolfsson, Bharat Chandar and Ruyu Chen. It draws on ADP payroll records covering roughly one in six American workers across more than seven hundred occupations. This is administrative reality rather than a survey of how anyone feels.
Workers aged 22 to 25 in the most AI-exposed occupations now sit about 19% below their peers in less-exposed occupations. That gap was 13% a year earlier. It is widening.
The single sharpest documented decline is among young software developers. Accounting and auditing are named among the occupations particularly susceptible.
Those are the same two occupations Gen Z named. A group with no research budget, no payroll data and considerably fewer options than any employer reached the same conclusion the administrative records show.
I want to be careful with what this proves, because most commentary on it is overclaiming. Stanford’s own headline reads “No Widespread Displacement.” The authors are explicit that these are descriptive patterns, not evidence that AI caused them. If you tell a board that AI has destroyed entry-level jobs, someone who has read the paper will correct you.
What the data does support is narrower and still worth your attention. Entry-level hiring in AI-exposed occupations is shrinking relative to everywhere else, and the shrinkage is happening through reduced hiring rather than layoffs. Nobody is being let go. Employers are simply not opening the door.
That distinction matters enormously, and it is why this has stayed invisible. Layoffs produce announcements, press coverage and questions at the AGM. A quiet decision not to run the graduate intake this year produces nothing at all.
Except that the people who would have applied notice immediately.
The contradiction you will be hit with
Now for the part that will get thrown at you if you repeat any of this in public.
In September, Fortune reported that Gen Z is flooding back into accounting, drawn by stability and clear six-figure paths. AI has taken the drudgery off the job and made it look attractive again.
So which is it? Are they fleeing white-collar work for the trades, or arriving in accounting in numbers we have not seen for years?
Both. And if you assert either one on its own, someone will produce the other and you will look like you picked the survey that suited you.
The mistake is treating a generation as though it makes one decision. It does not. What it does is sort, and sorting looks like contradiction from the outside.
Some are leaving fields they think will not hold. Others are entering fields where the pay is real and the pathway is visible. Those are not opposing instincts. They are the same instinct, which is to find work that will still exist in a decade and will pay for a house.
That is not disloyalty. It is the most conventional ambition there is.
Why it keeps getting read as flightiness
Here is the part I think we have backwards.
We have spent five years describing this generation as flighty, uncommitted and unwilling to pay their dues. Every survey about tenure gets reported as a character finding. They will not stay. They job-hop. They lack patience.
But look at what they are actually doing. They are researching AI exposure by occupation. They are weighing student debt against apprenticeship pay. They are watching which fields quietly stopped hiring, and adjusting.
That is not impatience. That is somebody pricing in a risk that their employers have not yet acknowledged out loud.
And this is the uncomfortable bit. When an organization stops hiring juniors, it rarely announces the decision or explains the reasoning. It shows up as a headcount line. But the twenty-two-year-old scrolling job boards sees it with total clarity, because they are the ones looking at the roles that are no longer posted.
They know. They have simply not been told, which means they have to guess at what it means, and they are guessing conservatively. Anyone would.
So when we call it a loyalty problem, we are describing an information problem and blaming the person who was kept in the dark about it.
What this changes for you
If you employ young people, or want to, this reframe has a practical edge.
The pitch that works on someone doing this kind of analysis is not culture, and it is not perks. It is durability. What will I be able to do in five years that I cannot do now, and who here is going to help me get there?
If you can answer that specifically, you will get people. If your answer is a values statement and a ping-pong table, you will lose to a plumbing apprenticeship, and you will call it a generational attitude problem rather than a competitive loss.
There is a harder question sitting underneath this one, which is what happens to your own leadership pipeline if the bottom rung of your organization keeps thinning. That is worth its own conversation and I will come back to it next week.
For now I would settle for this. The next time someone at your leadership table describes this generation as unwilling to commit, ask what they think a twenty-four-year-old is looking at that the rest of us are not.
They are reading the same map we are. They are just reading it more carefully, because they have further to travel on it.
What are the young people in your organization seeing that you are not?
#FutureOfWork #leadership #AI
If you’re leading an organization through this: the question worth asking isn’t why young people leave. It’s what they can see from where they’re standing that you can’t see from where you are.
If you’re planning a conference or leadership program: this is one of the conversations I have on stage, alongside live AI demonstrations built for the room in front of me.
Either way, send me a message. I read them all.
