The AI CV Problem

Every recruiter has felt it over the past two years: CVs that read a little too smoothly, cover letters with the same rhythm, the same three verbs, the same confident but strangely empty phrasing. Generative AI has become a routine part of how people apply for roles, from graduate applicants to seasoned executives.

A Workforce That Now Writes With Machines

The scale is significant. SHRM estimates that somewhere between 40 and 80 percent of applicants now use tools such as ChatGPT to draft resumes, write cover letters, or prepare for interviews. Resume Now’s 2025 survey of 925 US hiring professionals found that 57 percent had seen a noticeable rise in AI-assisted applications over the previous year, and 94 percent had encountered misleading or fabricated content produced by AI.

Harvard Business Review put it starkly in June 2026. In “AI Has Broken Hiring. Here’s How to Fix It”, Shraddha Sunil and Mudit Saraf describe how generative AI is undermining the reliability of signals recruiters have relied on for decades. Drawing on interviews with 120 talent-acquisition leaders and an analysis of more than 6,000 screening sessions, they warn that organisations risk selecting candidates who are best at navigating the hiring process, rather than those best equipped to do the job.

The New Reflex: Screen Out, Reject Fast

Employers have responded predictably, with more automation. Close to 98 percent of Fortune 500 companies now run candidates through an applicant tracking system, and a growing number layer AI-detection tools on top, scoring CVs for the tell-tale rhythm of machine-generated text. Resume Now found that 78 percent of companies now actively check for AI-generated content, and 62 percent say an AI-written CV that reads as generic and impersonal is more likely to be rejected outright.

The instinct is understandable. Nobody wants to shortlist a fabricated track record, and generic, keyword-stuffed CVs make a recruiter’s job harder, not easier. But building a hard “auto-reject” rule around an AI-detection score creates a different problem, one that boards and hiring teams are only beginning to reckon with.

Why the Detectors Are Wrong Almost as Often as They’re Right

AI-detection tools are built to spot patterns in continuous prose. CVs are the opposite of that: compressed, bullet-driven, keyword-dense documents that were never designed to carry a detector’s “voice fingerprint” in the first place. Industry analysis of detection-tool accuracy puts the false-positive rate on genuinely human-written material as high as 30 to 50 percent — high enough that a meaningful share of authentic CVs are being wrongly flagged, and quietly discarded.

The pattern of who gets caught in that net should concern any board serious about the quality and diversity of its leadership pipeline. One 2026 industry analysis of detection tools found non-native English speakers, who tend to write in a more formal, structured register, flagged at roughly 23 percent, against 4 percent for native speakers. Candidates with advanced degrees, whose academic training produces exactly the structured, polished prose that detectors mistake for machine output, are flagged at nearly double the rate of other applicants.

Because most of this filtering happens before a human ever opens the file, the candidate never learns why they were rejected. They simply disappear from the process.

What This Means at Senior and Executive Level

The stakes are different again once the conversation moves into the boardroom or the C-suite. Most senior candidates are not fabricating a career with AI. They are using it, much as they would use an editor or a coach, to sharpen how a genuinely strong track record is presented. Treating that CV the way a graduate-level ATS treats a thousand near-identical applications is a mismatch of tool to task.

At this level, the candidate pool is already thin, the search cycle is long, and a single missed appointment carries a cost measured in years, not weeks — echoing the warning boards have heard before about the price of getting a senior hire wrong. An automated reject on a false positive does not simply inconvenience one candidate. It can quietly remove the strongest person for the role from consideration before a client ever knows they existed.

Where Judgment Still Beats the Algorithm

This is precisely where a search partner earns their fee. At Jenova Partners, we do not screen candidates against an AI-detection score, and we do not let a CV, however it was drafted, stand in for the conversation. Every candidate we put in front of a client has been assessed through direct dialogue, rigorous referencing, and sector-specific scrutiny of what they actually delivered, not just how well it reads on the page.

That combination matters more, not less, as AI reshapes how CVs are written on one side of the desk and how they are screened on the other. Our role is to sit between those two automated processes and apply the judgment neither can: distinguishing a well-drafted account of real achievement from a well-drafted account of very little, and making sure the right candidate is never filtered out simply because a machine mistook their writing for another machine’s.

An algorithm can flag a CV in seconds. It takes a search partner who knows your sector to tell you whether the person behind it can actually do the job.

Sources

1. Shraddha Sunil and Mudit Saraf, “AI Has Broken Hiring. Here’s How to Fix It”, Harvard Business Review, June 2026.
2. SHRM, “Are Job Seekers Cheating When They Use ChatGPT to Craft Resumes and Cover Letters?”
3. Resume Now, “AI Is Flooding Hiring — But 62% of Employers Reject Resumes That Lack a Personal Touch,” 2025 AI and the Applicant Report (survey of 925 US HR professionals).
4. Jobscan, 2025 Applicant Tracking System usage research.
5. WasItAIGenerated, “AI Detection in Hiring: Screening AI-Written Resumes & Cover Letters,” industry analysis, May 2026.

Jenova Partners Limited is a private limited company registered in England and Wales (Company No. 17240640), with its registered office at 128 City Road, London EC1V 2NX.

Disclaimer: This article is intended for general informational purposes and does not constitute legal, financial, or investment advice.