That is the uncomfortable part of auto apply abuse.

The problem is not that candidates use AI to improve a resume. Many real candidates do that.

The problem starts when applications arrive at machine speed, from repeated devices, suspicious browser setups, proxy networks, and patterns that look coordinated.

A recruiter sees 300 applications.
The ATS says the funnel is healthy.
But the review queue tells a different story.

Too many submissions feel copied, low intent, or impossible to trust.

CrossClassify helps recruitment platforms and hiring teams detect auto apply candidate fraud through behavior, device, browser, network, velocity, and graph signals.

Not to reject candidates automatically.
To flag suspicious activity, explain the risk, and support human review.

I broke down the exact approach in the article linked below.

Read the full article here.