The Resume Is Dead. What Does That Mean for Screening?

The Resume Is Dead. Why do your best candidates sound exactly like your worst ones?

When everyone’s resume is AI-polished to perfection, the document that’s supposed to filter candidates becomes the thing hiding them.

A recruiter I know told me about a candidate last month. Stellar resume. Perfect keywords. Five years at a Series B startup that matched the role exactly. She passed the ATS, made it to the phone screen pile, and got scheduled.

Turns out the startup didn’t exist. Neither did the candidate—at least not as described. The person on the Zoom call couldn’t answer basic questions about the role they’d supposedly held for half a decade.

This isn’t an outlier anymore. It’s the new normal.

The numbers are getting worse

A 2025 Resume Builder survey found that 44% of job seekers admitted to lying during the hiring process. That’s just the ones who admitted it. Business Insider puts the broader number at 72% who’ve lied on a resume at least once. And 75% of hiring managers say they’ve caught a lie.

But here’s what’s changed: catching lies used to be about spotting inconsistencies. A date that didn’t add up. A title that seemed inflated. Now? AI writes resumes that are internally consistent, keyword-optimized, and grammatically perfect. The tells are gone.

Gartner predicts that by 2028, 25% of job candidates globally could be fake—driven largely by AI-generated profiles. We’re not talking about minor embellishments. We’re talking about synthetic identities, deepfaked video interviews, and coordinated reference networks where multiple fake personas vouch for each other.

The arms race nobody wins

Here’s the uncomfortable truth: we’ve built a system where candidates use AI to write resumes, and companies use AI to screen them. Both sides are optimizing for keywords and formats rather than actual capability.

The result? Everyone’s resume looks qualified. The signal-to-noise ratio has collapsed.

Recruiters now spend an average of 6-8 seconds on each resume. Not because they’re lazy—because they’re drowning. A single job posting averages 250 applications. Some roles in competitive markets see 500+. When you’re looking at hundreds of AI-polished documents that all hit the same keywords, you’re not screening for talent. You’re screening for whoever gamed the system best.

And ATS software isn’t helping. These systems reject 75% of resumes before a human ever sees them. One study found that ATS rejected three of a company’s own top five engineers when their resumes were submitted through the system. The filter has become the bottleneck.

What a resume can’t tell you

A resume is a marketing document. It’s supposed to be. But that means it’s optimized to present the best possible version of someone—and now AI can make that version indistinguishable from a genuinely qualified candidate.

What a resume can’t fake: how someone thinks through a problem in real time. Whether they can explain their experience in their own words. How they respond to follow-up questions. Whether they actually understand the work they claim to have done.

These are things that only come out in conversation.

The voice screening shift

This is why more companies are moving the conversation earlier in the process. Instead of resume → ATS → phone screen, they’re flipping it: application → voice screen → human review.

The logic is simple. If you talk to every candidate—even briefly—you learn things a resume will never tell you. You hear hesitation when someone claims expertise they don’t have. You notice when answers are rehearsed versus genuine. You catch the candidate who sounds great on paper but can’t string two coherent sentences together about their actual work.

AI voice screening makes this scalable. Instead of a recruiter spending 15 minutes on every phone screen (or more realistically, only screening the top 10%), an AI interviewer can have a structured conversation with every applicant. Not a chatbot asking yes/no questions—an actual dialogue that adapts based on responses.

The output isn’t a hire/no-hire decision. It’s a prioritized list of candidates worth the recruiter’s time, with conversation summaries that surface red flags and highlights. The recruiter still makes the call. They just make it with better information.

Trade-offs worth naming

Voice screening isn’t perfect. Some candidates get nervous. Some have accents that older speech recognition struggles with (though this has improved dramatically). Some genuinely qualified people don’t interview well.

These are real concerns. But they’re the same concerns that apply to human phone screens—and at least with AI, every candidate gets the same questions asked the same way. No bad days. No unconscious bias based on a name or voice.

The bigger issue is that voice screening only works if you’re actually listening for the right things. A poorly designed voice screen is just as useless as a poorly designed resume filter. The questions matter. The evaluation criteria matter. The handoff to humans matters.

Where this leaves us

Resumes aren’t going away. They’re still useful as a summary document, a reference point, a record. But they’re no longer a reliable first filter—not when AI can generate a perfect one in thirty seconds.

The companies adapting fastest are the ones treating the resume as one input among many, not the gatekeeper. They’re adding skills assessments, work samples, and yes, voice conversations to the top of the funnel.

If your current process is: post job → collect resumes → ATS filter → phone screen the top 5% → hope you didn’t miss anyone good—you’re probably missing good people. A lot of them.

Voice screening won’t solve hiring. Nothing will. But it adds a layer of signal that resumes used to provide and no longer do. In a world where the document is AI-generated, the conversation is what’s real.

If your team is buried in applications and wondering who’s actually worth talking to, RoundOne AI runs the first conversation with every candidate. 14 days free, no credit card.