This week a ranking made the rounds: America’s 20 most AI-resistant jobs, scored from most protected to least. Registered nurses came in first. Home health aides, elementary school teachers, and nursing assistants filled out the top four. Software engineers landed in the middle of the pack. Wholesale sales reps came in dead last.
People read that list and drew the obvious conclusion. The safe jobs are the low-paid ones. The exposed jobs are the ones that pay well. Learn to mop a floor, I guess, and skip the degree.
That conclusion is wrong, and it’s wrong in a way that matters if you run a business or manage a team.
The List Isn’t About Money. It Never Was.
Pull the wage data next to that ranking and the pattern falls apart fast. A home health aide, the second most AI-resistant job on the list, makes around $17 an hour. A software engineer, sitting in the middle of the pack at fifth, makes closer to $60 an hour on average. If wage and AI-resistance were connected, that gap shouldn’t exist. It does, because they’re not measuring the same thing.
What actually predicts resistance is whether the work happens in a body, in an unpredictable place, in front of another human who needs something from you right now. A nurse reading a patient’s face. A repair worker crawling into a crawlspace that doesn’t match the blueprint. A server at a good restaurant reading the table and knowing when to disappear and when to check in. None of that is about skill level or paycheck. It’s about whether a machine can physically do the job yet. Right now, mostly, it can’t.
Meanwhile the work that gets hit hardest is information processing, regardless of how much it pays. Bookkeeping. Scheduling. First-draft writing. A lot of what a $130,000-a-year software developer does on a given Tuesday. McKinsey’s research backs this up directly: this wave of automation is different from the last one. Past automation came for manual labor. This one is going straight after office and knowledge work, the stuff that happens on a screen.
The Junior Developer Problem
Here’s where it gets specific enough to actually matter. Stanford’s Digital Economy Lab tracked real payroll data and found something split right down the middle of a single profession. Entry-level software developers, ages 22 to 25, saw employment drop nearly 20% since late 2022. Developers 30 and older, doing the exact same job title, grew employment 6 to 12% over that same stretch.
Same occupation. Same list ranking. Completely different outcome, based on what specific tasks each group was actually doing. The junior devs were writing boilerplate and running basic tests, exactly the kind of work AI tools do well today. The senior devs were making judgment calls, architecture decisions, the stuff nobody’s built a tool for yet.
That’s the real lesson buried in this list, and it applies far past software. The job title doesn’t tell you who’s safe. The task does.
Where Do the Displaced People Actually Go?
This is the question the viral list skips entirely, and it’s the one that matters to you as an owner. Say a role in your business gets automated. Where does that person go?
History gives an answer, and it’s not “nowhere.” When automation wiped out corporate librarian jobs, it also created search engine optimizer and web designer jobs, roles that didn’t exist yet. Workers moved. Not instantly, and not without pain, but the shift wasn’t a dead end. Sectors that made up 3% of employment in 1910, mostly healthcare and professional services, make up almost 30% today.
The World Economic Forum projects something similar this time: 92 million roles displaced globally by 2030, and 170 million created. That’s a net gain on paper. It’s also a different set of people doing the losing and the gaining.
Here’s the number that should actually get your attention: workers with AI skills right now earn a 56% wage premium over workers without them. A year ago that premium was 25%. It more than doubled in twelve months. That gap won’t close. It’ll widen while people wait for a clearer explanation of what’s happening.
What This Means For Your Team, Not Just The Economy
A U.S. Chamber of Commerce Foundation survey from earlier this year found that among small businesses already using AI, 73% say it’s changed what their employees actually do day to day. 65% say it’s changed who they hire. That’s not a future trend. That’s this year, in businesses your size.
So look at your own team, one task at a time, not one job title at a time. Somebody on your payroll is doing the digital equivalent of writing boilerplate code. Data entry. Scheduling. First drafts of emails and posts. That work is exposed no matter what the job title on their business card says.
And before you point at everyone else’s job title, check your own week. If half of it is bookkeeping, scheduling, and first drafts, you’re not exempt from this list either. You just don’t have a rank number yet.
The move isn’t to panic and it isn’t to pretend it’s not happening either. It’s to plan the direction each person on your team needs to move. Up, meaning they build judgment and skill that sits above what AI can do. Sideways, meaning they pick up a different skill entirely, the way the librarian became the web designer. Down is rarely the answer, but pretending nobody has to move at all is worse than any of the three. Nobody puts “learned something new sideways” in their LinkedIn headline. Doesn’t matter. It still works.
Start The Conversation This Week
You don’t need a training budget or a twelve-month plan to start this. Pick one person on your team whose role leans heavy on repetitive, screen-based tasks. Sit down with them this week and ask what part of their job they’d actually want to get better at, if the boring half got automated.
Most owners never have that conversation until the decision gets made for them. You have a year or two of runway here, not overnight. Use it.
Here is the link to the original article. https://www.visualcapitalist.com/cp/ranked-americas-most-ai-resistant-jobs/