How to Spot Fake Job Applicants at the Screening Form (2026)
Recruiters now spend hours a week on applications that were never real: fake identities, mass-submitted AI resumes, duplicates. The screening form is the cheapest place to catch them, if it asks the right things.
Two different problems
"Fake applicants" covers two things that need different responses. The first is fraud: invented identities, people applying on someone else's behalf, and in remote roles, candidates who plan to have someone else do the interview or the job. Fraud vendors such as Sardine report this growing in tech and finance hiring.
The second is volume. Tools now let one person send a tailored application to 200 jobs in an afternoon. The applicant is real, but the application isn't much of a signal: they may barely know what your role is. A LinkedIn report found that more than half of recruiters who said hiring had got harder blamed the rise in AI-generated applications.
A screening form handles both, as long as it asks for things that are cheap for a real candidate and expensive to fake at scale.
Checks that cost the candidate nothing
- Temporary email domains. Mailinator, 10-minute mail and similar. A real candidate who wants a reply gives a real address.
- Fake phone numbers. 9999999999, 1234567890, the same digit repeated. Easy to check automatically.
- Duplicates. The same answers submitted twice, or a long answer that matches another applicant's almost word for word. Near-duplicates across different names usually mean one template used for many fake identities.
- Location that doesn't agree. A stated city, a phone area code and a time-zone availability that point to three different places.
- Repeat screen-out attempts. Someone who fails your knockout questions, then tries again from the same device with different answers until they qualify.
Screening questions that are hard to fake
Generic questions get generic answers. "Why do you want to work here?" is one prompt away from a perfect paragraph. Ask things that need specific knowledge of the work or of the person's own past.
- "Describe a time a deploy went wrong. What broke, and what did you change afterwards?"
- "What tool do you use for X, and what's one thing about it that annoys you?"
- "Our support queue gets about 300 tickets a day in two languages. What would you look at first in your first week?"
- "Which of the requirements in this job post do you not meet yet?"
A real candidate answers these in two minutes with something only they would say. A mass applicant either skips them or pastes something that reads like a brochure. That difference is what you're looking for.
What not to penalise
Be careful with the shortcuts. Polished writing is not fraud; many strong candidates now use AI to tidy their wording, and penalising that mostly penalises careful people. Non-native English is not a signal. A gap in a work history is not a signal. Recruiting platforms such as Breezy make the same point: flag copy-paste and stock content, not AI use as such.
Flag, sort, then decide
Automated checks should change the order you review applications in, not who gets reviewed. Put the strongest-evidenced applications first and the flagged ones last, with the reasons visible. In the EU, recruitment AI is classed as high-risk under the AI Act, which brings requirements for human oversight. Keeping a person on every rejection is both the fair approach and the safe one.
How SiliForm helps
On every plan, SiliForm checks each answer as the candidate types it. An answer that doesn't address the question gets a short reason and a chance to try again, and temporary email addresses are rejected outright. A hidden spam trap and a rate limit stop basic bots.
On the Pro plan, every application gets a quality score from 0 to 100 with the reasons shown: duplicate or near-duplicate answers, fake phone numbers, rushing, repeated screen-out attempts from one device, answers that contradict each other, and open answers that read as pasted from an AI tool. You can sort and filter by Clean, Review and Flagged, change any label by hand, and export with the quality columns included. Nothing is rejected automatically.
See SiliForm for recruiters, or start from the candidate screening form template.
Frequently asked questions
What are the signs of a fake job applicant?
Common signs are contact details that don't hold up (temporary email domains, VoIP or patterned phone numbers), a location, time zone and phone area code that all disagree, a resume that doesn't match the candidate's public profile, the same unusual phrasing across several applications, and answers that contradict each other. One sign is a reason to look closer; several together are a reason to verify before interviewing.
How can I reduce spam job applications?
Put a short screening form in front of the resume upload, with two or three role-specific questions that need a concrete answer. Block disposable email domains, check for duplicate submissions, and score responses so recruiters review the best-evidenced applications first.
Is it legal to screen applicants automatically?
Rules vary by country. The EU AI Act treats AI used in recruitment as high-risk, which brings requirements around human oversight. A safe pattern is to use automated checks to flag and sort, and keep the decision to reject with a person. Check local employment law before you deploy any screening.
Can SiliForm help screen job applicants?
Yes. AI validation rejects off-topic or nonsense answers and temporary email addresses while the candidate fills in the form, on every plan. On Pro, each application gets a quality score with reasons, covering duplicates, fake phone numbers, speed, contradictions and AI-written answers. Nothing is rejected automatically; recruiters see the flags and decide.