70% of Hiring Managers Trust AI. Only 8% of Candidates Agree.

Trevor Higgs | April 2026

The most dangerous number in talent acquisition right now isn't a budget figure or a time-to-fill metric. It's 62. That's the gap [3], measured in percentage points, between how hiring managers feel about AI in hiring and how candidates feel about it.

70% of hiring managers say they trust AI to help make better hiring decisions. Only 8% of candidates agree. And that gap isn't narrowing. It's accelerating.

The Canyon Nobody Talks About

For years, the HR technology conversation has focused on efficiency. How many applications can we screen? How fast can we move candidates through the funnel? How much recruiter time can we save?

Those are valid questions. But they've created a blind spot: nobody asked the candidates how they felt about the process.

The answer, it turns out, is terrible. 52% of workers say they're more worried than hopeful about AI in the workplace. 75% don't feel confident using AI tools. And when it comes to AI [4] making decisions about their careers, trust levels drop to single digits.

This isn't a perception problem you can solve with a better careers page. This is a systemic failure of transparency.

The Glassdoor Effect

Every hiring decision now leaves a digital footprint , and not just in your ATS.

Rejected candidates don't send polite feedback forms. They write Glassdoor reviews. They start Reddit threads. They post TikTok videos that rack up thousands of views. They drop warnings in professional group chats and alumni networks.

"I spent 45 minutes talking to a camera and got rejected by an algorithm."

"No one could explain why I wasn't selected."

"The whole process felt like a surveillance experiment."

These aren't isolated complaints. They're becoming the dominant narrative about AI-powered hiring. And every one of those posts is being read by candidates who haven't applied yet.

Your employer brand isn't what your careers page says. It's what rejected candidates tell their friends.

Why the Gap Keeps Widening

The trust gap is widening because the industry has prioritized two things: efficiency and "innovation" , often at the expense of the one thing candidates actually care about: being treated like a human being.

Consider what "advanced" assessment technology has looked like over the past five years: facial micro-expression analysis, vocal pattern scoring, eye-movement tracking, and mouse-behavior monitoring. The pitch to employers was compelling: more data points, better predictions, faster decisions.

The experience for candidates? Invasive, confusing, and often indistinguishable from surveillance.

When a major vendor halted their facial analysis product after an FTC complaint, it wasn't a surprise to candidates. They'd been saying it felt wrong for years. The industry just wasn't listening.

What Trustworthy AI Actually Looks Like

The path to closing the trust gap isn't paved with better PR. It's paved with better methodology.

Trustworthy AI in hiring has four non-negotiable characteristics.

1. Explainability. Every decision must be explicable in plain language , not corporate jargon, not statistical notation, plain language that a candidate can understand. If you can't explain why a candidate wasn't selected in one clear sentence, your methodology has a transparency problem.

2. Respect. No cameras, no voice analysis, no biometrics, no behavioral surveillance. Assessment should measure what predicts performance, not what creates discomfort.

3. Human control. AI should recommend. Humans should decide. The moment you remove humans from the loop, you've created a system that candidates cannot appeal , and a system that cannot appeal is a system that cannot be trusted.

4. Scientific validation. Claims about predictive validity should be backed by peer-reviewed research, not marketing materials. The Schmidt, Oh, and Shaffer meta-analysis spanning 100 years of research [5] is clear: general cognitive ability (r = .65) is the single strongest predictor of job performance. It's 4x more predictive than years of experience and 2.5x more predictive than reference checks.

The Business Case for Trust

The trust gap isn't just an ethical issue. It's a business problem with measurable costs.

Companies with strong employer brands see 50% more qualified applicants. They reduce cost-per-hire by up to 50%. They cut turnover by 28%. But those numbers only hold if candidates trust your process.

When the trust gap widens, your talent pool shrinks. Not because candidates aren't out there , because they're choosing competitors whose processes don't feel like surveillance.

In a labor market where 41% of frontline [6] workers turn over annually and early-career hiring has dropped 73% [7], you cannot afford to lose candidates to a trust problem.

The Path Forward

Closing the trust gap requires a fundamental shift in how we think about assessment technology. Not as a tool for efficiency, but as a product whose users are the candidates themselves.

Would you ship a product that 92% of your users don't trust? Of course not. But that's exactly what most AI hiring tools are doing.

The companies that win the talent war in 2026 won't be the ones with the most sophisticated AI. They'll be the ones with AI that candidates can understand, explain to their friends, and feel respected by.

Trust isn't a feature. It's the whole product.

[1] Greenhouse. (November 2025). 2025 AI in Hiring Report. Survey of 4,136 respondents across US, UK, Ireland, and Germany.

[2] Greenhouse. (November 2025). 2025 AI in Hiring Report. Survey of 4,136 respondents across US, UK, Ireland, and Germany.

[3] Derived from Greenhouse. (November 2025). 2025 AI in Hiring Report. Calculated as difference between 70% hiring manager trust and 8% candidate trust.

[4] Pew Research Center. (February 25, 2025). U.S. Workers Are More Worried Than Hopeful About Future AI Use in the Workplace. Survey of 5,273 U.S. adults conducted October 7-13, 2024.

[5] Schmidt, F. L., Oh, I.-S., & Shaffer, J. A. (2016). The Validity and Utility of Selection Methods in Personnel Psychology: Practical and Theoretical Implications of 100 Years of Research Findings. Working paper. (Original meta-analysis: Schmidt, F. L., & Hunter, J. E. (1998). Psychological Bulletin, 124(2), 262-274.)

[6] iCIMS. (2025). 2025 Q4 Frontline Hiring Report.

[7] Ravio. (2025). 2025 Tech Job Market Report. Entry-level positions (P1 and P2 job levels) saw a 73% decrease in hiring rates.