By Varun Modgill ยท August 31, 2026
Everyone Looks Qualified Now. That's the Problem.
A resume used to be hard to write. That was its core value. Tailoring one took hours, proving a candidate understood the role and cared enough to do the work. You weren't screening the document, you were screening the effort behind it.
AI removed that effort. The document survived; the signal died.
Volume Is the Symptom. Uniform Perfection Is the Crisis.
LinkedIn now processes roughly 11,000 applications per minute, up 45% year on year. But volume is just a symptom. The real crisis is uniform perfection: when AI generates polished, keyword-optimized resumes in seconds, documents no longer separate talent. This quickly escalates from resume polish to identity risk.
- Global threat: Gartner projects 1 in 4 candidate profiles worldwide will be fake by 2028.
- Local impact: In June 2026, Infosys deferred assessments for 20,000+ applicants after detecting widespread impersonation.
AI Detectors Are the Wrong Fix
Knee-jerk buying of AI detectors backfires in non-native English markets. A Stanford study revealed two fatal flaws:
- Severe bias: Detectors falsely flagged 61% of essays written by non-native speakers, misinterpreting careful, second-language phrasing as AI-generated text.
- Negative signal only: Detectors only identify candidates to reject; they never reveal whom to hire.
The lasting solution is simple: stop screening for things that are costless to fake.
Separate the Artifact from the Evidence
Hiring teams must separate two concepts they have historically treated as identical. Artifacts submitted to your process, such as resumes, cover letters, and self-declared skill lists, are candidate-authored, unobserved, and free to generate. Their signal value has collapsed.
Conversely, evidence generated inside your hiring process requires real effort and cannot be fabricated. That internal evidence is now the only defensible foundation for a hiring decision.
From Self-Reported Claims to Verified Proof
Cognavi shifts your hiring pipeline from what candidates claim to what your process can prove:
- Institutional verification: Academic records and project histories enter through direct university partnerships, confirmed at the institution level.
- Patented skill tree: Structures verified data into a granular capability model, evaluating a candidate's placement in a skill graph rather than relying on stuffed keywords.
- Direct observation: Proctored assessments and video interviews capture real capability inside the platform.
The goal is not to eliminate friction, but to make faking a profile cost significantly more than simply being competent.
Reach Is Abundant. Proof Is Scarce.
For thirty years, hiring was about reach. Today, reach is noisy and abundant, but proof is scarce.
Stop drowning in immaculate resumes for candidates you cannot verify. Test a verification-first pipeline on your hardest open roles by booking a demo with Cognavi at https://www.cognavi.com/book-a-demo
Where this goes next
Graduate hiring for enterprisesSee how Skill Tree matching delivers a shortlist of verified engineering graduates instead of a keyword-filtered pile of resumes.
Written by
Varun Modgill
Managing Director, Cognavi India
Varun Modgill is Managing Director of Cognavi India, where he leads the skills-first hiring platform connecting engineering graduates, universities and enterprises across India and Japan. He writes on campus hiring, skills-based assessment and the graduate talent market.