All posts

Two-Sided Trust Collapse: How AI Screening Is Failing Both Recruiters and Candidates

Hiring used to run on a simple, if imperfect, contract. Candidates presented their best selves. Recruiters evaluated honestly, sometimes imperfectly, but in good faith. Both sides trusted that the process, however slow or flawed, was at least real. That contract is breaking down, and AI screening is caught in the middle of it.

Sep 14, 20263 min read

Two Different Stories About the Same System

Ask a candidate about AI screening and you'll often hear frustration: a black box that rejects them without explanation, that seems to reward keyword-stuffing over substance, that feels less human every year. Nearly half of job seekers say their trust in hiring has dropped over the past year, and AI is a major reason why. Ask a recruiter and you'll hear a different kind of frustration: a flood of applicants, some of whom are actively trying to game the system. The vast majority of recruiters and hiring managers say they've spotted or suspected candidates manipulating AI screening tools, including embedding hidden instructions in resumes designed to trick the software into flagging them as top candidates. Two groups. Two grievances. Same broken middle layer.

Why This Is a Trust Problem, Not Just a Technology Problem

It's tempting to treat this as a technical bug: patch the vulnerability, update the model, move on. But the deeper issue is structural. When candidates believe the system is arbitrary or unfair, some will conclude that gaming it is the only rational move. When recruiters can't tell a genuinely strong candidate from a manipulated result, they either over-trust a flawed system or stop trusting AI screening altogether, undermining the reason it was adopted in the first place. Each side's distrust reinforces the other's. Candidates who feel screened unfairly are more likely to try workarounds. Those workarounds make recruiters less able to trust the process. That erosion makes screening feel even more opaque to candidates. It's a feedback loop, not a one-time failure.

The Numbers Behind the Breakdown

  • The majority of hiring managers now believe candidate manipulation of AI tools is a real, active problem, not a hypothetical one.
  • A meaningful share of candidates openly admit to trying to manipulate AI screening directly.
  • Only a minority of HR leaders feel confident they have strong controls to catch this kind of manipulation.

Put together, this isn't a fringe issue. It's a system where the incentives on both sides are drifting away from honesty, and the safeguards haven't caught up.

What Restoring Trust Actually Requires

Fixing this isn't about picking a side: cracking down harder on candidates, or scrapping AI tools to appease job seekers. It requires closing the gap from both directions. For recruiters and HR leaders:

  • Build in verification layers that check for manipulation attempts, not just keyword matches
  • Be transparent with candidates about how screening works, even at a high level
  • Treat AI screening output as one input, not a final verdict

For the hiring process itself:

  • Design systems that reward genuine signal over gameable patterns
  • Give candidates a fair, legible path, reducing the incentive to look for backdoors
  • Audit screening tools regularly against known manipulation tactics

The Real Question

AI didn't create distrust in hiring. But it gave both sides a new reason to doubt the other, and a new set of tools to act on that doubt. The organizations that get ahead of this won't be the ones with the flashiest AI screening tool. They'll be the ones who treat trust as something to actively rebuild on both sides of the table, not something they can assume the technology will handle for them.