There’s a belief taking hold, in this industry more than most, that AI will eventually make recruitment obsolete — and that if AGI ever arrives, whatever’s left of the human role in hiring goes with it. It’s not a fringe view. Say it out loud in a room full of recruiters and you’ll get uncomfortable nods more often than pushback.
We don’t share it. Not because we’re betting against the technology — we use it every day, and we expect to use more of it, not less. We don’t share it because we think the belief rests on a category error: it assumes hiring is, at bottom, a computation.
Match the resume to the role, the skills to the requirements, the candidate to the culture — do that well enough, fast enough, and there’s nothing left for a person to add. If that’s what hiring actually is, the belief is probably right, and it’s just a matter of time.
We don’t think that’s what hiring actually is. And the reason takes a slightly different shape than the usual case for "the human touch."
What AI is built to do with a mistake
Start with what AI is actually built to do. Every serious AI system is engineered to be error-tolerant — to minimize mistakes, and when it makes one, to correct it from the outside: retrain the model, patch the weights, adjust the guardrails. The system itself doesn’t sit with the error.
It doesn’t feel the weight of having been wrong. Something else — an engineer, a process, a next version — absorbs the correction and hands back a system that behaves as if the mistake never happened.
What a person does with one instead
People don’t grow that way, and it’s not a limitation — it’s the entire mechanism. A person learns almost exclusively through their own mistakes and the lessons that follow them. Not a patched update handed down from outside, but a lived, often uncomfortable process: getting something wrong, feeling the consequence of it, and choosing — nobody makes you — to do better next time.
That choosing is where humility actually comes from. It’s where honesty comes from, the kind that admits a real gap rather than papering over it. And it’s where a genuine, self-directed desire to grow comes from — not a training objective someone else defined, but something the person decided for themselves, usually the hard way.
Why that matters in recruiting, specifically
That distinction matters more in recruiting than almost anywhere else, because the job is, at its core, reading that exact quality in another human being. Has this person actually grown from their setbacks, or have they just gotten better at avoiding situations where they’d have to?
Is the humility in the room real, or is it a well-rehearsed answer to an interview question they saw coming? Does the drive to improve come from inside them, or does it evaporate the moment nobody’s measuring it? Recognizing the real thing — the difference between a person who has genuinely been humbled by their own mistakes and one who has simply learned to perform humility — isn’t really pattern recognition.
It’s closer to recognition, full stop: you know it because you’ve lived a version of it yourself. That’s not a dataset. It’s a life.
A layer AI may not be able to answer
Which is where we land on the deeper claim, the one that’s harder to argue cleanly and the one we believe anyway: call it spirit, call it soul, call it whatever word survives the conversation without embarrassing anyone — there is a layer of a person that sits beneath and beyond whatever any system is optimizing for.
We don’t think that layer is something AI is currently missing and will eventually catch up to. We think it may be a question AI itself is not equipped to fully define, measure, or answer — not because the technology isn’t powerful enough, but because the question was never a computational one to begin with. That’s not a knock on the technology. It’s a claim about where the question actually lives.
Where this leaves us
None of this makes us anti-AI — quite the opposite. We’re AI-forward by design: we use it to cut through noise, surface signal faster, and clear the volume that used to eat a recruiter’s week. What we don’t do is let it make the final call on a person.
That’s deliberate, and it’s not nostalgia for how things used to be done. It’s a bet on what we don’t think changes, no matter how intelligent the tools become: that there will always be a need — if not an outright desire — for people to be evaluated, understood, and chosen by other people, especially at the moments that matter most in a career or a company’s direction.
That’s the belief this firm is built on, and it’s the reason the read on a candidate, at PHM, still ends with a person who has been wrong before, learned something from it, and is willing to bet their judgement on someone else’s next chapter.
PHM Search is an executive and specialist recruitment firm for mid-market engineering, construction, and technology companies across Canada and the USA. If you’re weighing a hire where judgement matters more than a keyword match, start a conversation.
