Credential fraud in remote hiring — what 1-in-6 means.

If you hire remotely, especially across borders, you’ve probably already encountered fraud. You might not have caught it. Most companies don’t. The 1-in-6 statistic that keeps showing up in the literature — the share of remote applicants for technical roles whose credentials don’t survive verification — is the public-facing tip of a much larger iceberg.

What the fraud actually looks like

It is no longer just "lied about a degree." The patterns we see in customer pipelines include:

• Stand-in interviews — the person on the screen isn’t the person on the CV. Particularly common in the Indian and Eastern European IT services market.
• Deepfake video — growing rapidly since 2024. Real-time face replacement is now within hobbyist budgets.
• Fabricated employment — LinkedIn profiles, references, even mock company websites set up for the duration of one application.
• Credential inflation — the easiest one. Two years of experience becomes five. A bootcamp becomes a CS degree.
• Identity rotation — one technically-skilled person, multiple identities, applying to fill multiple roles at once for outsourcing arbitrage.

Why ATSs miss it

Traditional ATSs treat verification as a downstream step — after the offer, sometimes after the start date. By then, the cost of catching fraud is enormous: terminations, rehiring, legal exposure, team morale. The fraud that makes it past your interview process has already cost you most of what catching it could save.

Verification at offer is too late. By then, you’ve already invested 60+ hours per hire and signalled internal commitment.

What stops it

Three things, in combination:

1. Video authenticity checks during the interview itself. Face consistency, lighting analysis, prompt-response timing. Real-time alerts to the recruiter when artifacts appear.
2. Credential cross-checking on every claim. Degree verification against the National Student Clearinghouse (US), UGC (India), equivalents in EU. Employment dates against public records where possible.
3. Cross-source identity correlation. Does the LinkedIn match the CV? Do the GitHub commits align with claimed employment? Do the timestamps make sense?

None of this should be a separate product. It should be part of the hiring platform itself, running silently in the background, surfacing only when something looks off. That’s what DeepTrust does inside CognitoHire.

What you owe your real candidates

The reason fraud detection matters isn’t mostly about catching the bad ones. It’s about respecting the good ones. Every fraudulent candidate who makes it through your funnel is a real candidate whose application got buried under the noise. Every undetected synthetic profile is a recruiter hour stolen from someone who deserved your attention.

Fraud detection isn’t adversarial security theatre. It’s a form of respect for the candidates who are genuinely showing up.