WHY CANGRADE
Candidate authenticity, from the first resume to the final reference.
Cangrade’s candidate fraud detection software ensures your recruiters spend their time on authentic candidates from the earliest point in hiring.
Candidate fraud detection that starts at the first touchpoint
Most candidate fraud detection tools start at the identity check or the interview. But by then, your team is already invested. Recruiters have read the resume, sent and reviewed assessments, scheduled the call, and spent the hour.
Cangrade’s candidate fraud detection starts inside Resume Screen, so questionable applications are flagged before anyone on your team spends time on them.
One platform · Three layers
Authenticity at every hiring stage
Cangrade goes beyond most candidate fraud detection tools, protecting the resume, the assessment, and references, all in the platform your team already uses.
01
Resume Screen
Flags AI-generated, inflated, and mirrored applications before a recruiter reads them.
02
Talent Assessment
Prevents cheating by design, so AI has no answer key to find.
03
Reference Check
Flags references connected to the candidate before an offer goes out.
01 · Resume Screen
What Cangrade catches at the resume screen
Passive candidate fraud detection software waits for something to look unusual. Cangrade’s proprietary method actively hunts for and detects fraud, so it has fewer places to hide.
AI-inflated resumes
Resumes the candidate started, then AI rewrote, overstating experience and skills along the way.
Fraudulent applications
Applications entirely fabricated by AI, not by the person whose name is on them.
Job description mirroring
Resumes that echo your posting, without the evidence of experience a qualified candidate brings.
Tailoring is smart. Mirroring isn’t experience.
Good candidates tailor their resumes to the role. Cangrade can tell the difference between a candidate describing real experience and a resume repeating your job description back. Our proprietary active fraud detection looks at:
- AI file and writing signals
- Candidate history
- Location verification (optional)
02 · TALENT ASSESSMENT
How Cangrade prevents cheating on assessments
Most assessments try to catch cheaters in the act. Cangrade’s behavioral pre-hire assessment is designed so cheating doesn’t work in the first place.
Each assessment measures soft skills against a success profile built for that specific role and organization, so an answer that’s strong for one job can be a red flag for another. Dynamic content with no right answers leaves AI nothing to look up. Copy-paste prevention, time limits, response-pattern flags, IP flagging, and question randomization add additional layers of protection.
BUILT-IN LAYERS
✓ No right answers
✓ Copy-paste prevention
✓ Time limits
✓ Response-pattern flags
✓ IP flags
✓ Question randomization
CANGRADE RESEARCH
We had ChatGPT take our assessment for 50 roles
It was instructed to maximize its score. Even with copy-paste prevention and time limits disabled, it scored as a high fit only 12% of the time.
Preventing cheating is smart. Adding hoops isn’t.
Many anti-cheating tools protect hiring tests at the candidate’s expense with webcam monitoring, screen recording, and lockdown browsers. That kind of experience feels like surveillance, and it shows up in your completion rates and employer brand.
Cangrade prevents cheating through how the assessment is built, not by watching candidates take it. The design does the work, so honest candidates get a short, respectful experience.
03 · REFERENCE CHECK
What Cangrade catches at the reference check
Reference fraud often looks like a friend or family member posing as a former manager or coworker. Cangrade looks for signs that a reference is connected to the candidate or isn’t a genuine professional contact, so your team can follow up before an offer goes out.
- IP duplication: flags when a candidate and a reference respond from the same IP address
- Employment verification: confirms candidates’ workplaces, positions, and length of employment
- Company email requirement: references respond from a work email address rather than a personal one
Every flag explained. Every decision yours.
Every authenticity flag comes with a plain-language explanation your team can review and defend. Flags reflect how an application was produced and the authenticity of its content, never who the candidate is.
Authenticity flags don’t change candidate scores. They add context for your recruiters, and the final decision always stays with your team.
Works with the tools you already trust
Cangrade’s candidate fraud detection software runs inside your existing hiring workflow and ATS. It complements identity verification and background checks by covering what they can’t see: whether the application itself reflects the candidate.
Proven in practice
10x
more accurate predictions of success1
81%
decrease in time-to-hire2
52%
average decrease in turnover3
98%
candidate satisfaction4
Common questions about Cangrade’s candidate fraud detection software
What is job description mirroring?
Job description mirroring is when a resume reproduces the language of a job posting without evidence of the experience it describes. It’s increasingly common because AI tools make it effortless to generate a resume that matches any posting. Cangrade flags mirroring for recruiter review, so qualified candidates who tailor honestly still move forward.
How is this different from identity verification or background checks?
Identity verification confirms a person is who they say they are. Background checks confirm past employment and records. Cangrade’s candidate fraud detection software answers a different question, earlier in the process: does this application genuinely reflect the candidate’s experience? The three work best together.
Will candidates who tailor their resumes be flagged?
No. Candidates who tailor their resumes in their own words won’t be flagged. Flags are reserved for applications that rely on AI to create content or documents, or that mirror the posting without supporting evidence.
Can candidates use ChatGPT to pass Cangrade’s assessment?
Not reliably. When our research team had ChatGPT take the assessment for 50 roles, it scored as a high fit only 12% of the time, even with copy-paste prevention and time limits set aside.
Does Cangrade automatically reject flagged candidates?
No. Flags go to a recruiter with an explanation and don’t change candidate scores. Cangrade’s AI doesn’t make the hiring decisions, your team does.
How does Cangrade keep fraud detection fair?
Flags are based on how an application was produced, never on demographic characteristics. Every flag is explainable and reviewed by a person, consistent with Cangrade’s commitment to fair, explainable hiring.
1Cangrade internal study
2Cangrade customer data
3Cangrade customer data
42025 Trustpilot rating