What changed in applicant screening
The old screening problem was scarcity: too few qualified candidates and too much manual sourcing. The new problem is mixed volume. A single job can attract hundreds of applications, and many of those applications are now produced or polished by AI tools. Some are honest candidates using technology to write more clearly. Others are low-quality submissions that copy the job description, exaggerate experience, or invent credentials.
That distinction matters. Using AI to improve grammar is not the same as fabricating employment history. Recruiters need a process that separates normal writing assistance from material misrepresentation. A blanket ban on AI-written resumes is nearly impossible to enforce and can punish candidates who simply needed help presenting real experience. A verification workflow is more practical and fair.
Signals that deserve closer review
A suspicious application is rarely obvious from one signal. The pattern matters. A disposable email domain on its own may simply mean a candidate is privacy-conscious. A profile link on an unexpected domain might be a typo. A resume that matches the job description closely may be a motivated applicant. But when several signals appear together, the recruiter should slow down and verify before advancing.
Common signals include disposable email domains, profile links that impersonate LinkedIn or portfolio sites, unsafe link schemes, repeated application metadata across unrelated candidates, copied application answers, unusually similar CV text, and employment timelines that do not hold up against the candidate profile. None of these proves fraud. They are reasons to review the application with more care.
The key is to avoid turning a warning into a verdict. The system should say, "review this," not "reject this." That protects candidates and keeps hiring decisions accountable.
Why fair review beats automatic rejection
Automatic rejection based on integrity signals is risky. Data can be incomplete, links can be mistyped, and legitimate candidates sometimes use privacy tools, temporary addresses, or sparse online profiles. If a system silently rejects those people, the hiring team may never know it excluded a qualified candidate for the wrong reason.
A fair process keeps humans in control. The ATS can highlight a warning, explain what triggered it, and put the candidate in front of the recruiter for manual review. The recruiter can then compare the CV, application answers, interview notes, references, and work samples. That is a stronger decision than relying on a black-box score.
A practical workflow for suspicious applications
Start by reviewing the warning in context. Check whether the signal is about contact data, profile links, duplicated content, or application behavior. Next, verify the underlying claim. If the issue is a LinkedIn mismatch, compare dates and titles. If the issue is a suspicious portfolio link, ask the candidate to confirm the correct URL. If the issue is repeated text, use the phone screen to probe for detail.
The best questions are specific. Ask the candidate what problem they solved, what tradeoffs they considered, who they worked with, what failed, and what they would do differently. Fabricated experience often sounds confident in general terms but lacks operational detail. Real experience usually includes context, constraints, and imperfect outcomes.
How Treegarden frames the problem
Treegarden treats candidate integrity as an advisory workflow. The warning icon on the candidate card is a prompt for manual review, not a fraud label. The candidate details panel explains the warning in plain language so the recruiter can decide what to verify next.
This is deliberately different from a "fake applicant detector" that claims certainty. In hiring, certainty is rare. The safer product decision is to show the recruiter where to look, keep the signal explainable, and preserve human oversight for every candidate decision.
For teams already using AI-generated resume detection checklists or application flood management workflows, integrity warnings add another layer of triage without replacing recruiter judgment.
Review applications with context
Treegarden helps recruiters manage high-volume pipelines with advisory AI, application integrity warnings, and human review built into the hiring workflow. Book a demo
Frequently Asked Questions
Does a warning mean the candidate is fake?
No. A warning means the application contains one or more signals that deserve manual review. It should never be treated as proof of fraud or used as the sole reason to reject a candidate.
What is the safest way to handle suspicious applications?
Review the signal, verify the claim, document the decision, and keep a human recruiter responsible for the outcome. Avoid fully automated rejection for integrity warnings.
Should companies ban AI-written resumes?
A blanket ban is difficult to enforce and can be unfair. A better policy is to allow writing assistance while requiring every claim about experience, credentials, and achievements to be accurate and verifiable.