71% of People Don’t Want AI Making the Hiring Call. Small Teams Can Automate Everything Else.

The Race to Hand Over the Decision

The direction of travel is unmistakable. Recruiters already lean on AI to sift the bulk of incoming résumés, and the next wave is agentic: tools that don’t just assist but act — sourcing, messaging, screening, and ranking candidates end to end, with more than half of talent-acquisition leaders saying they plan to deploy that kind of system this year. Each step gets framed as a natural extension of the last. If AI can read the applications, why not have it score them? If it can score them, why not let it decide who advances?

The logic feels airtight from inside the funnel, where the problem is volume and the metric is time saved. But it quietly moves the line from “AI helps me do the work” to “AI makes the judgment,” and those are not the same thing. One clears busywork. The other decides someone’s livelihood. Treating them as a single slope is how a reasonable efficiency drive ends up somewhere the people it affects never agreed to go.

Candidates Are Voting the Other Way

When the Pew Research Center asked U.S. adults about this directly, the split wasn’t close. A striking 71% opposed using AI to make a final hiring decision; just 7% were in favour. Opposition softened the further you got from the decision itself — on merely using AI to review applications, 41% were against and 28% in favour — but the pattern held: the closer AI gets to the verdict, the harder people push back.

And this isn’t just an opinion people hold in the abstract. In the same research, about two-thirds of adults said they would be less likely to apply for a job if they knew AI was being used to help make the hiring decision. Sit with that for a second. It means the “let the machine decide” approach doesn’t only risk a worse decision — it thins the pipeline before you ever get to make one. The reasons people gave were not anti-technology; they were specific. AI can’t capture everything about a person. It misses context. It can carry its own biases while wearing the costume of objectivity. These are not paranoid objections. They are a fair description of what a model does and doesn’t see.

Why the Backlash Hits Small Teams Hardest

A large employer can absorb some candidate resentment. The brand is known, the perks are real, and there is always another cohort of applicants. A small team has no such cushion. When you’re hiring your fifth or fifteenth person, every strong candidate who self-selects out because your process felt like an algorithm graded them is a genuine loss you can’t easily backfill. The people most likely to walk are exactly the ones with options — the ones you most wanted.

But flip that around and it’s the whole opportunity. If two-thirds of candidates are wary of faceless, automated decisions, then being visibly, genuinely human is a differentiator that costs a small team almost nothing to offer and is very hard for a big automated pipeline to fake. A real person read your application. A real person will make the call and can tell you why. For a company without a famous logo, that’s not a soft nicety — it’s one of the few structural advantages you actually have. The trap is throwing it away in the name of efficiency, automating the one part of hiring that candidates most want to stay human.

Draw the Line at the Decision

None of this is an argument against using AI to hire. The volume is real, and reading every résumé by hand is not a viable plan for a team of one or two. The distinction that matters isn’t AI versus no AI. It’s where you point it. Aim it at the work — the sorting, the scheduling, the note-taking, the first pass that surfaces who’s worth your time — and it buys back the hours that let a human do the judging well. Aim it at the verdict, and you spend your scarcest asset, candidate trust, to save time you didn’t need to save.

That line is exactly where we built Kynto to sit. It takes the volume off your plate — turning a role into clear criteria, screening applications against them, and handling the scheduling and coordination that eat a solo recruiter’s week. Its meeting bot joins the interview and produces a speaker-attributed transcript, so nothing rides on half-remembered notes. But when it comes to the decision, it deliberately keeps a human in the seat: after each interview it asks the interviewer a structured question set tied to the role’s criteria, and it builds the candidate’s score primarily from that human feedback, cross-checked against what was actually said — rather than letting a model quietly decide on its own. You get the speed of automation and a decision you can stand behind and explain to the candidate. You can see how that works at kyntoai.com.

Key Takeaways

• Candidates draw a hard line at the decision. In Pew Research Center polling, 71% opposed AI making a final hiring call — and about two-thirds said they’d be less likely to apply to a job that uses AI to help decide.
• For a small team, that backlash is a cost you can’t absorb — the candidates most likely to self-select out are the ones with options. But being visibly human is also an edge a big automated pipeline can’t easily copy.
• The useful line isn’t AI versus no AI — it’s where you point it. Automate the work (sorting, scheduling, transcripts, the first pass) and keep the judgment human, so the decision is one you can explain.

The companies that handle this well won’t be the ones that automated the most. They’ll be the ones that were clear-eyed about what to automate and what to protect. The application flood is a genuine problem, and AI is a genuine part of the answer — pointed at the work. But the moment it’s pointed at the verdict, you’re trading away the trust of the exact people you’re trying to hire. Small teams don’t have to make that trade. Keeping a person in the seat for the decision was always the better way to hire. Now it’s also what candidates are asking for out loud.