Why 78% of Candidates Prefer Voice AI to Humans

Correction, 3 September 2026: an earlier version cited “80% of candidates” from unnamed studies and a “3 to 4 times” completion figure. The preference figure is 78%, from the University of Chicago Booth field experiment, whose offer, start and retention results replace the unsourced numbers. If you assumed candidates would resist AI-conducted interviews, the data disagrees.

In a University of Chicago Booth field experiment that randomly assigned about 70,000 applicants to a human recruiter or an AI voice interviewer, 78% of candidates given the choice picked the AI. That’s not a rounding error. It’s a consistent signal about something real in how candidates experience the hiring process.

Here’s what’s driving it — and what it means for your hiring outcomes.

Why Candidates Choose AI

Candidates aren’t naive about what they’re interacting with. They know it’s an AI. They prefer it anyway. The reasons are practical and psychological:

Judgment-free environment. In a human interview, candidates are hyper-aware of being assessed on non-verbal cues, accent, appearance, and enthusiasm level — factors they know are proxies for bias. With a voice AI, candidates report feeling assessed on what they say, not how they present.

Available on their schedule. A voice AI doesn’t have a calendar. Candidates can complete their screening at 10 PM after their current job ends, or at 6 AM before their shift. This matters enormously for employed candidates — the highest quality cohort.

No small-talk anxiety. The ritual social performance of a human interview — the handshake equivalent, the weather comment, the self-deprecating opener — is exhausting for many candidates. Voice AI removes this overhead entirely.

Consistency. Candidates who have experienced vastly different human screeners — one friendly, one terse, one distracted — report that AI consistency feels fairer. Same questions. Same tone. No mood-dependent outcomes.

What the Numbers Show

The preference for voice AI translates directly into measurable outcomes:

The 18% lift in job starts is the most operationally important figure. When you send a batch of candidates to a human phone screen, many never schedule, many reschedule, and some ghost entirely. Voice AI invitations see dramatically higher uptake — because there’s no scheduling friction and no social anxiety about “bothering” a recruiter.

The 12% lift in job offers is the most strategically important figure, and the study’s authors attribute it to better signal extraction: the AI interview surfaces information that predicts real match quality. The AI screening is part of the employer brand experience.

The Deloitte Signal

Deloitte’s 2026 HR technology predictions say voice interfaces will be the primary accelerant of AI adoption across all HR functions this year. The reasoning: voice is the natural interface for assessment. It captures what matters — communication, articulation, confidence, coherence — in a format that’s native to human communication.

Text-based chatbots and form-filling assessments feel like homework. Voice conversations feel like conversations. The completion rates reflect this.

The HireQwik Conversation Design

HireQwik’s screening conversations run 15–20 minutes — calibrated specifically based on candidate dropout data. Shorter than 12 minutes doesn’t collect enough signal. Longer than 25 minutes sees completion rates fall.

The conversation structure follows what enterprise HR professionals told us they actually want to know:

1. Communication quality first — can this person express themselves clearly?
2. Role comprehension — do they understand what this role actually involves?
3. Domain basics — do they have the foundational knowledge the JD requires?

Candidates rate this conversation structure highly because it feels purposeful. They’re not being asked trick questions or given brain teasers. They’re being asked to talk about the role they applied for. That’s the interview they wanted.

What This Means for Your Candidate Pool

There’s a second-order effect that HR teams often miss: voice AI expands your reachable candidate pool.

Top candidates — the ones with options — are the most likely to reject high-friction screening processes. They won’t record a video submission. They won’t complete a 45-minute async assessment. They have other offers to consider.

But they will spend 15–20 minutes on a voice conversation, because it’s low friction, schedule-flexible, and — critically — it feels respectful of their time.

The employers who deploy voice AI screening don’t just process existing applicants faster. They attract a qualitatively different applicant pool.

The Candidate Experience Is the Employer Brand

In a market where candidates share hiring experiences publicly — on LinkedIn, on Glassdoor, in peer networks — the quality of your screening process is a public signal about your organisation.

A well-designed voice AI screening sends the message: “We’ve thought about your time. We’ve built a process that’s fair, structured, and human-like in the right ways.”

That’s not a small thing when you’re competing for the same talent pool as everyone else.

Want to give your candidates a better first impression? Start a pilot and run your next screening batch through HireQwik. See the completion rates yourself.