The AI Jobs Debate Is Missing the Most Important Point

Type: Opinion / Discussion Driver

Everyone is arguing about which jobs AI will take.

Radiologists. Paralegals. Copywriters. Customer service agents. The predictions range from sober (some roles will shrink) to apocalyptic (half of all jobs, gone within a decade).

It's a reasonable thing to worry about. But I think it's the wrong question — and focusing on it is causing people to miss something more important.

The question worth asking is this: which skills become dramatically more valuable in a world where AI can do a lot of things pretty well?

The answer changes what you should be doing right now.

What the Displacement Debate Gets Wrong

The history of automation is not a history of permanent mass unemployment. It's a history of tasks being automated and humans moving to different tasks — some predictably, some not.

When ATMs spread across the US in the 1970s and 80s, most economists predicted bank teller jobs would collapse. They didn't. The number of bank tellers actually increased. Cheaper bank operations meant banks opened more branches, which needed more tellers — but those tellers spent less time on cash handling and more time on customer relationships and financial advice.

This doesn't mean AI won't displace workers — it will, and it already is. But the displacement tends to be of specific tasks within jobs, not jobs wholesale. And the pattern is consistent: routine, predictable, definable tasks get automated. Judgment, relationship, and novelty get more valuable.

Which Skills Compound Alongside AI

Here's what I think the evidence points to.

Taste and judgment. AI can generate a hundred options. Knowing which one is actually good — and being able to explain why — is a human skill that hasn't been automated. Curators, editors, creative directors, senior engineers reviewing AI output: all of these roles get more valuable as the volume of generated content increases.

Communication. AI is a fluent writer of competent prose. But the ability to communicate complex ideas persuasively, to adapt to a room, to navigate difficult conversations — these remain distinctly human. If anything, the baseline for written communication is rising, which means genuinely excellent communication stands out more.

Domain expertise combined with AI fluency. A lawyer who understands AI tools deeply can do in hours what used to take days. A doctor who can interpret and interrogate AI diagnostic support is more valuable than one who can't. The combination of deep expertise and AI fluency is rare right now — which makes it extremely valuable.

Relationship and trust. AI cannot be your doctor's face on a difficult day. It cannot be the manager who believed in you when you didn't believe in yourself. The parts of work that are fundamentally relational are not going away.

What History Tells Us About Tech Transitions

The Industrial Revolution didn't eliminate work — it transformed it. Millions of agricultural jobs disappeared, but millions of factory jobs appeared, then millions of service jobs. The transition was often brutal for individuals while being net positive at a societal level. That's cold comfort if you're in the middle of it.

The honest answer about AI's impact on employment is: we don't know the full picture yet. The technology is moving faster than our ability to measure its second-order effects. Anyone who tells you with certainty that it will all be fine, or that it will be catastrophic, is not reading the evidence carefully.

What I believe — and you can push back on this — is that the people who will navigate this transition best are not the ones trying to protect the tasks they do today, but the ones learning how to work with AI as a tool while developing the skills that remain distinctly human.

Your Concrete 3-Step Plan

First: audit what you actually do each week. Separate the tasks that are definable and repeatable from the ones that require your specific judgment, relationships, or creativity. The first category is at risk. The second is not.

Second: get fluent with AI tools in your field. Not because they're good at your job, but because understanding their capabilities and limits makes you a better judge of their output — and makes you more valuable to employers navigating this shift.

Third: invest in your distinctly human skills. The ability to communicate, to lead, to think strategically, to build trust with other humans. These are not at risk. They are becoming premium.

I said at the start that everyone's asking the wrong question. So let me ask the right one:

What do you do that an AI genuinely couldn't do — even a very good one? Think about it. Your answer is probably your most valuable professional asset right now.

Do you agree? I'm genuinely curious where you'd push back on this. Tell me in the comments.