The trust gap no one is talking about.
AI adoption in hiring is accelerating. Candidate trust is not.
A recent poll conducted by the Staffing Industry Analysts found that 63% of job seekers have now been interviewed by AI. At face value, that signals rapid normalisation. But dig one layer deeper and a more significant finding emerges: 38% of candidates have walked away from a hiring process specifically because it involved an AI interview.
What that number reflects is not resistance to technology. It reflects a trust deficit, and trust deficits in hiring processes have measurable consequences for pipeline quality, employer brand, and legal exposure.
What candidates are actually telling us
When a candidate exits a process, they rarely do so quietly. They tell their network. They leave reviews. In a talent market where employer reputation shapes pipeline quality, candidate withdrawal is a signal that carries operational weight.
What the data points to is a distinction that organisations are still learning to make: the difference between AI in a process and AI that can be explained within it.
Candidates are not categorically opposed to being assessed by intelligent systems. Many prefer structured, consistent evaluation over the variability of under-prepared human interviewers. What they are opposed to is being assessed without context, without understanding, and without recourse.
A related problem sits alongside the transparency issue. When every candidate for a role receives an identical set of questions, regardless of their background, experience, or what their CV actually says, the process communicates something unintended. It signals that the system is not engaging with them as individuals, but processing them as a batch. That perception erodes confidence in the fairness of the outcome, even before a result is delivered. A genuinely intelligent assessment adapts. It draws on what the candidate has already told you, follows the thread of their experience, and asks the questions that are actually relevant to them specifically.
When a process feels like a black box and sounds like a script, it is reasonable that candidates start to question whether the outcome will be fair.
The business cost of unexplained AI
For HR and Talent Acquisition leaders, the implications run in two directions.
First, there is the pipeline risk. If a meaningful percentage of candidates disengage at the AI screening stage, the quality of the shortlist is already compromised before a single hiring manager conversation takes place.
This is compounded by the broader volume challenge: application numbers have grown by as much as 239% across many roles since the widespread adoption of generative AI tools, with 79% of job seekers now using AI in their application process. The pool is larger. The signal is harder to read. And if your screening process drives away engaged candidates, the problem compounds further.
Second, there is the legal and governance exposure. Under the EU AI Act, UK GDPR, and EEOC guidance, candidates increasingly have the right to understand how automated systems influenced decisions about them. Unexplained AI decisions create audit risk. Where documentation is absent and reasoning is unavailable, organisations are exposed, both to regulatory challenge and to reputational harm.
Transparency is structural. It needs to be built into the process from the outset, not retrofitted after a complaint or an audit request.
Explainability as competitive advantage
This is where the conversation shifts from risk mitigation to strategic positioning.
Organisations that can tell candidates, with specificity, how they are being evaluated will attract more engaged applicants. Structured, explainable processes signal that an organisation takes hiring seriously, values consistency, and respects the people moving through its pipeline.
That signal matters at every stage: in job adverts, in screening communications, in the interview itself, and in how decisions are communicated.
When candidates understand how they are being assessed, they trust the process more. When they trust the process more, they engage with it more fully. The output, the quality of evaluation data your team receives, improves directly as a result.
What this means for 2026 hiring strategy
The organisations that will lead on talent acquisition over the next three years are those that deploy AI responsibly, visibly, and with candidate experience built into the design of every process from the start.
That means asking a harder question of every tool in your stack:
Can we explain what this system does, and why, to every candidate who moves through it?
If the answer is uncertain, the risk is already present. With 70% of recruiters reporting significantly higher pressure in 2026 to move faster while maintaining quality and fairness, the case for structured, explainable AI has never been more operationally clear.
The bottom line
Candidate withdrawal from AI-assisted processes is a signal worth taking seriously. When the reasoning behind an assessment is unclear, trust erodes, and with it, engagement, pipeline quality, and employer reputation.
Explainability serves multiple functions simultaneously: it satisfies regulatory requirements, it supports defensible decision-making, and it makes the hiring process more credible to the people moving through it. In a hiring landscape where candidate trust is under pressure, that credibility has real strategic value.
Sources
- Staffing Industry Analysts - Candidates Have Some Trepidation About AI in Hiring, Poll Says (2025). staffingindustry.com
- The Economist (2026), reporting application volume increases of up to 239% across roles since the rise of generative AI tools. Also: HireWithLumi research.
- LinkedIn Global Talent Trends Report (2026); Gartner CHRO Priorities Survey (2026).
