Resume Fraud Detection: How to Verify What a Resume Claims
Back to Blog

Resume Fraud Detection: How to Verify What a Resume Claims

By Keara Garrett. Last updated: November 3, 2026.

Resume fraud detection means checking each claim on a resume against a source the candidate does not control. That covers past employers, schools, certification bodies, references and a live test of the skills listed. Most fake resumes fall apart at one of those checks, not at a first read.

This guide is for HR, talent acquisition and procurement teams hiring tech and AI consultants in the US and Canada. It covers what to check, how to check it, and when in the process each check belongs.

What counts as resume fraud?

Resume fraud is an intentional misstatement meant to mislead the person hiring. Researchers Christine Henle, Brian Dineen and Michelle Duffy split it into three types in a 2019 study in the Journal of Business and Psychology:

  • Fabrication. Stating something false, like a degree never earned or a job never held.
  • Embellishment. Stretching something true, like a bigger title or a larger role on a project.
  • Omission. Leaving something out on purpose, like a short job that ended badly.

The same study found resume fraud predicted lower job performance and more workplace deviance after hire. So it is more than a paperwork problem.

In practice, the claims worth checking fall into five groups:

Claim What fraud looks like Who can confirm it
Employment Invented employers, stretched dates, inflated titles The past employer
Education Unearned degrees, degree mills, changed majors or dates The school's registrar
Certifications Expired, never issued or borrowed credentials The issuing body
Skills Tools or methods listed but never used A live test or work sample
References Friends posing as managers The reference's employer

A candidate profile can be honest and still polished by AI. A profile can also be handwritten and false. This guide covers checking whether claims are true. For AI-written candidate profiles, see How to Spot AI-Generated Resumes.

How do you detect fake resumes at the screening stage?

At screening, you are looking for claims that do not fit together. A red flag is a reason to verify, not a reason to reject. Plenty of honest careers look odd on paper.

These are worth a closer look:

  • Dates that clash. Two full-time roles that overlap, or a degree finished while working full time across the country.
  • Years without months. "2021 to 2023" can hide a gap of almost two years.
  • Employers you cannot find. No website, no business registration, no other staff online.
  • Titles that do not fit the company. A "Head of AI" at a five-person agency is possible, but worth asking about.
  • Tools that did not exist yet. A framework listed for a role that ended before the framework was released.
  • Certifications out of order. An advanced credential listed before the one it requires.
  • A profile that disagrees with itself. Dates or titles that differ between the resume, LinkedIn and the application form.
  • References that are hard to place. Personal email addresses, or no way to confirm the person worked there.

Write down what you noticed. It tells the person running the next check what to look at first.

How do you verify employment history?

Confirm each employer exists, then confirm the candidate worked there, using contact details you find yourself.

  1. Check the company exists. Search its website and a business registry. In the US, that is the Secretary of State business search for the state. In Canada, use Corporations Canada's federal corporation search for federal companies or the provincial registry.
  2. Find the main line yourself. Call the company's published number, not the one on the resume. Ask for HR or the candidate's manager.
  3. Confirm the basics. Exact title, start and end dates, and whether the role was full time, part time or contract.
  4. Ask about rehire eligibility. Some employers will answer this even when they share nothing else.
  5. Write it down. Record who you spoke to, when, and what they confirmed.

Many employers will only confirm dates and title. That is still useful. It tells you the role was real, and the skills check covers the rest.

For consultants who worked through an agency or their own company, ask who the end client was. Then confirm with both the agency and the client where you can.

How do you verify degrees and certifications?

Check that the school is real and recognized, then confirm the degree with the school itself.

Step United States Canada
Is the school legitimate? Search the Department of Education's Database of Accredited Postsecondary Institutions and Programs Search CICIC's Directory of Educational Institutions in Canada
Did the candidate graduate? Contact the registrar, or use a degree verification service the school works with Contact the registrar directly
Degree from another country? Use a credential evaluation service Use a credential assessment service; CICIC lists them

Check the details, not just the degree. The field of study, graduation date and credential level should all match the resume. A graduation year that overlaps a full-time job in another city is worth a question.

For certifications, go to the issuer. Most major cloud and AI certifications have a public verification page or a digital badge you can check. Confirm three things:

  • The certification was issued to this person.
  • It was valid during the time the resume claims.
  • It has not expired, since many require renewal.

Never rely on a certificate file the candidate sends you. A PDF is easy to edit.

How do you verify the skills on a resume?

Have the candidate show the skill live, with someone who knows the work in the room. Paperwork can confirm a job existed. Only a test shows what the person can do.

  • Pick two or three claims to test. Choose the skills the role depends on most, not everything listed.
  • Run it live. A take-home task is easy to hand to someone else. A live session or a live walkthrough of the take-home is not.
  • Ask them to explain a past project. What was the goal? What did they personally build? What went wrong? Real experience comes with detail and trade-offs.
  • Separate their work from the team's. "We shipped the model" is fine. Ask what part was theirs.
  • Review work samples together. Code, notebooks or architecture diagrams are useful only if the candidate can walk you through them.

For AI and data roles, good questions cover the messy parts. Ask how they handled bad data, how they judged whether a model was good enough, and why they chose one approach over another. These are hard to answer from a script.

How do you check that references are real?

Confirm who the reference is before you trust what they say. A fake reference is often a friend with a convincing script.

Before the call:

  • Check the reference held a role where they would have worked with the candidate, at the same company and time.
  • Prefer a work email on the company's domain over a personal address.
  • Where you can, reach them through the company's main line rather than the number you were given.

During the call, ask for specifics a stranger could not invent:

  1. In what role did you work with this person, and for how long?
  2. Tell me about one project you worked on together. What was their part?
  3. What were they strongest at? What would you want them to improve?
  4. Would you hire or work with them again?

Vague praise, no project detail and answers that echo the resume word for word are all reasons to dig further.

When should each check happen?

Run light checks early and the slow, formal ones near the offer. That way no one waits weeks for a background check on a candidate who would not pass the skills test.

Stage Check Why here
Screening Compare dates, titles and tools across resume, LinkedIn and application Free, fast, and catches the clearest problems
First conversation Ask the candidate to walk through two roles in their own words Gaps between paper and speech show up quickly
Technical interview Live skills test with a subject expert Tests the skills the role depends on
Finalist Employment, education and certification checks; reference calls Worth the time once the shortlist is small
Offer Background check and work eligibility, with the candidate's consent Formal checks with legal rules attached

Scale the depth to the role. A senior consultant with access to client data or production systems warrants every step. A short, low-risk engagement may not.

How do you stay fair and legal in the US and Canada?

Get consent first, check only what the role needs, and apply the same checks to every candidate for the role. The rules differ by country.

United States Canada
Main law The Fair Credit Reporting Act (FCRA), when a third party runs the check. State and local rules can add more. PIPEDA for many private employers. Alberta, British Columbia and Quebec have their own private-sector privacy laws.
Before the check A standalone written disclosure and the candidate's written authorization Meaningful consent, with the purpose explained
What you can check What is relevant to the job What is reasonable and related to the role
Before rejecting on the results A pre-adverse action notice with a copy of the report, then a final adverse action notice Let the candidate respond to what you found

The FTC and EEOC publish a joint plain-language guide, Background Checks: What Employers Need to Know. In Canada, the Office of the Privacy Commissioner publishes Privacy in the Workplace guidance. This section is a summary, not legal advice. Check your process with counsel.

Watch for false positives. Honest candidates can trip a red flag. Career changers, people with international careers and consultants who worked through agencies often have records that are harder to confirm. Before you reject anyone, tell them what did not match and give them a chance to explain.

How does Augmnt check consultants?

Augmnt is a marketplace that connects businesses with verified AI, data and tech consultants. Every consultant is checked before their candidate profile reaches you, so your team starts from a verified shortlist rather than a pile of unverified resumes.

Each consultant goes through four checks:

Check What it covers
Resume fraud Our platform scans each resume for warning signs: dates that do not add up, wording that mirrors the job description too closely, and patterns common in AI-written text. A flag is a reason to look closer, not a verdict. A person then confirms the listed employers are real and that the resume matches the consultant's other public profiles.
Skills We check the certifications and skill badges a consultant lists on their public profiles. Then we interview them closely on the skills the work needs, to confirm they know how to do it.
Identity and location We confirm each consultant is a real person based in the US or Canada, using their phone number, email and interviews.
Background review We review each consultant's public professional history and compare it with what we learn in interviews.

Your own interviews and final checks still matter. Verified candidate profiles mean those steps confirm fit, instead of starting from zero.

Browse verified AI and data consultants on Augmnt, or have our team source and deliver them for you.

Frequently asked questions

How common is resume fraud?

No one knows the true rate. Most widely shared figures come from self-reported surveys, often run by companies that sell screening tools, and they vary a lot. One useful data point: in a July 2025 Gartner release, 6% of 3,000 job candidates surveyed admitted to interview fraud. Gartner also predicts that by 2028, one in four candidate profiles worldwide will be fake. That is a forecast, not a measurement.

Is it illegal to lie on a resume?

Not automatically, but it is risky. Depending on what the lie is about and the role, it can amount to fraud.

In Canada, a material lie that the employer relied on when hiring can be just cause for dismissal without notice, because it breaches the employee's duty of honesty. Simple exaggeration usually is not enough. (Canadian HR Reporter) In serious cases, faking credentials to gain work can fall under the Criminal Code's fraud section.

In the US, a material lie can also justify firing. If the lie causes the employer real damage, the employee can face a lawsuit. Rules differ by state and province.

Can software detect a fake resume on its own?

Software can flag inconsistencies, like clashing dates or mismatched profiles. It cannot confirm that a job, degree or skill is real. That still takes the employer, the school, the issuing body or a live test.

Is using AI to write a resume the same as resume fraud?

No. Using AI to tidy wording is not fraud. Using it to invent jobs, projects or skills is. Check the claims, not the writing style. More in How to Spot AI-Generated Resumes.

How long does resume verification take?

Screening checks take minutes. Employment and education checks depend on how quickly employers and schools respond, so start them as soon as a candidate reaches the final round.

Do I need to verify contractors and consultants too?

Yes. Contractors often get access to systems and client data quickly, sometimes faster than employees. Apply the same checks, scaled to the access they will have.