Everyone's Hiring for AI Skills. The Smart Companies Are Hiring for Judgment.

(What a new HBR survey gets right about the talent question underneath the AI question)

A CEO and her head of people asked me to sit in on a planning session last quarter. The company's about 140 employees, growing fast, and AI had landed on the agenda the way it does at every company that size — somewhere between exciting and faintly threatening. The CEO opened with the question she'd clearly been carrying for weeks:"Do we need to go hire AI people? Everyone keeps telling me we're behind."

Her head of people, who's sharp and not easily rattled, answered before I could. "I think we need to figure out what we're actually trying to do first. I'm not sure 'AI people' is even what we're short on."

She was right. And a lot of companies are quietly arriving at the same conclusion.

The number that surprised me

Harvard Business Review Analytic Services surveyed 230 small and midsize businesses late last year. Most findings are what you'd expect — 79% agree AI is driving the need to upskill existing talent, plus the usual anxiety about training budgets.

But here's the one that stopped me. Asked who they'd hire if they could pick only one person, 52% chose a candidate with deep relevant industry experience, 34% wanted an equal mix of industry and AI knowledge, and only 7% prioritized a candidate with strong AI experience.

Sit with that. The companies most worried about falling behind on AI, given the choice, overwhelmingly picked the person who knows the business over the person who knows the technology.

That's not fear of the technology. That's wisdom.

What's actually going on here

Let's untangle this, because it's easy to misread. These companies aren't rejecting AI — three-quarters plan to increase their use of it this year. What they're discovering, often the expensive way, is that AI skills without judgment are close to worthless.

The researchers have a name for what these companies are circling toward: the "wisdom economy," where organizations value workers for their ability to discern AI's flaws and capabilities. They call the people who can make those calls "wisdom workers." As one expert in the report put it, we spend a lot of time asking what AI can do, when we should be asking what we want it to do. That second question isn't technical. It's human.

Why this matters more for you than for the giants

Here's the genuinely good news for growth-stage companies, and it runs against the usual narrative.

The big enterprises can outspend you on AI talent — multimillion-dollar offers, signing bonuses, the works. You can't win that auction, and you shouldn't try. But the skills they're paying a fortune for — prompt engineering, model-building, gen AI infrastructure — are the ones that commoditize fastest. What doesn't commoditize is someone who understands your business deeply enough to know where AI helps and where it quietly breaks things.

And that person is far more likely to be in your company than in a giant one. Seventy percent of those surveyed said AI is driving the need for talent with the creativity, intuition, and discernment to work with it — exactly the qualities you can't buy off a résumé. You grow them. You usually already have them, in people who've been told their whole careers that their "soft" skills were the nice-to-have.

They were never the nice-to-have. The market just took a while to catch up.

We've seen this movie before

If this feels like a leap of faith, it isn't. A well-known Harvard Business Review study by Ranjay Gulati, Nitin Nohria, and Franz Wohlgezogen tracked 4,700 public companies across three recessions. Of those that survived, only 9% came out genuinely stronger — outperforming rivals by at least 10% on sales and profit growth. The interesting question is what that 9% did differently.

Not a single lever. They mastered a balance: cutting costs selectively, focusing harder than rivals on efficiency, while still investing in the future instead of slashing everything in a panic. The pure-defense companies survived but stalled; the pure-growth ones overextended. The winners had the judgment to tell the difference — when everyone around them was reacting instead of thinking.

That's the same muscle this AI moment asks for. The companies that win the next few years won't be the ones that spent the most on AI. They'll be the ones with the judgment to invest in the right things while competitors lurch between panic-buying AI talent and ignoring it entirely. The technology changes. The discipline of knowing where to point it doesn't.

What I'd actually do Monday morning

Notice that almost none of this involves a job posting.

Ask the "want," not the "can," question first. Before anyone evaluates a tool, get your leadership team to answer: what do we actually want AI to do, and what do we want to stay stubbornly human? That surfaces your real talent needs faster than any skills audit — and the report found 56% of these companies expect difficulty just figuring out which AI skills they need.

Look down before you look out. The instinct under pressure is to hire your way out. But your best wisdom workers are probably already on payroll — the people who know why a customer is really calling, why a deal actually stalled. Give them room to play with the tools. That costs you permission, not budget.

Stop treating judgment as ambient. Most companies invest in technical skills and assume judgment just accumulates. It doesn't — not at this speed. Discernment, ethical clarity, knowing when to overrule the confident-sounding machine: these are developable, and right now they're your scarcest resource. Put them in your development plans, right next to the tooling.

The work underneath the work

This is most of what I do, honestly. I partner with CEOs and people leaders at growth-stage companies on exactly this layer — not "should we adopt AI" (you will), but the harder talent optimization questions underneath: who here has the judgment we're going to lean on, how do we develop more of it, and how do we keep the wisdom workers we already have from quietly leaving while we chase the ones we think we need.

Sometimes that's alongside an HR leader, sometimes before there is one. It usually starts with the realization that CEO had in the planning session — that the question wasn't "do we need AI people." It was "do we know what we're trying to become."

If any of this is landing, I'd love to hear what you're seeing.

Survey data from "The New Talent Playbook for Small and Midsize Businesses in the Age of AI," Harvard Business Review Analytic Services, 2026. Recession research from Ranjay Gulati, Nitin Nohria, and Franz Wohlgezogen, "Roaring Out of Recession," Harvard Business Review, March 2010.