AI has made it trivial to blast hundreds of applications in minutes. Inbound pipelines are filling up with noise: fake profiles, bot submissions, and applications that bear no resemblance to the person behind them.
The volume problem was already hard enough. Now there's a quality problem layered on top, and manual triage misses the patterns that flag a fake.
So we built the fix. Fraud detection is now built directly into Application Review. Every application is automatically assessed for signs of fraud at the point of review, with no extra setup and no separate tool.
Fraud detection now in Application Review
What fraud detection looks for
Fraud detection evaluates every application across two dimensions, both running automatically the moment an application lands in Application Review.
Identity deception asks whether the candidate is who they say they are. We cross-reference the details on an application against a range of signals to assess whether the person behind it matches their claimed identity.
Application automation asks whether the application was submitted by a bot. We look for the telltale patterns of mass-apply tools and automated submissions, so genuine interest is easier to separate from inbox spam.
The detection logic is deliberately conservative. Multiple signals must compound before fraud detection surfaces a risk level on a candidate, so no single weak signal triggers a flag. That keeps false positives low and your attention on the candidates who deserve it.
How it shows up in your workflow
The whole point is that the work happens before you ever see the application. Recruiters don't run a separate pass to check for fraud. Fraud signals sit alongside the application, the resume, and your AI evaluation criteria, so the inbound recruiting workflow you already have just gets sharper.
- Fake profiles slip into the shortlist alongside real candidates.
- Bot submissions clog the queue and crowd out the people you actually want to read.
- Recruiters spend time on applications that were never going to be genuine in the first place.
- Identity-deception signals are surfaced at the point of review.
- Application-automation patterns are flagged before bot submissions reach your shortlist.
- Each flagged application carries a clear risk level and a short explanation of why.
High-risk candidates are flagged automatically before they move further in your process. Each flag comes with a risk level and a brief explanation, so the recruiter has the information to decide quickly without leaving the inbox. Compare it to running fraud checks as a separate pass after the screening calls have already eaten the day.
See it in action
Here's what a flagged candidate looks like inside the Application Review inbox. The risk indicator sits next to the candidate's row, the reasoning expands inline, and the same Reject and Progress actions you already use stay one click away.
- 1The flag itself, naming the dimension that fired (identity deception, application automation, or both).
- 2The compounded risk level, so the recruiter sees the strength of the signal at a glance.
- 3The same Reject and Progress actions, available the same way they were before.
As with everything else in Application Review, the system surfaces the signals and you make the final call. Fraud detection is there to give a recruiter the information they need to act quickly, not to make the decision for them.
Availability
Fraud detection is on by default for every Application Review customer, with nothing to switch on and nothing in Settings to configure. The first flagged candidate is the first time the feature shows itself, and that's intentional.
Every fake application that makes it through is wasted time for your team and a worse outcome for the real candidates competing for the same role. Fraud detection is one more step toward making sure the people on your shortlist actually deserve to be there.
Frequently asked
How does fraud detection decide an application is risky?
A single weak signal is never enough on its own. Multiple indicators across identity and submission patterns have to compound before a flag fires, which is what keeps false positives low and protects genuine candidates whose application might have one unusual but innocent trait.
Do I need to turn fraud detection on?
No. It's running for every Application Review customer the moment it lands, with no toggle in Settings and no opt-in flow. If your team isn't yet on Application Review, that's the prerequisite step. See our pricing for what's included.
What happens to a candidate that gets flagged?
Flagged candidates stay in the pipeline with a visible risk level and short explanation. From the inbox, the recruiter can Reject them, Progress them anyway, or open the application to read the underlying signals before deciding. The flag is information, not an automatic action.
Does this work with my ATS?
Yes. Fraud detection runs on every application that lands in Application Review, regardless of which ATS feeds it. Greenhouse, Ashby, and the rest of the supported ATS list all feed in the same way, with new integrations shipping on the same cadence.
Bring Metaview into your hiring stack.
Live notes, structured scorecards, and ATS sync - set up in under 10 minutes.
