Identity Fraud Prevention at Onboarding: Deepfakes, Fake IDs and Duplicate Accounts
Stop account opening fraud before the account exists. In one KYC session we check the ID document for tampering, confirm a live person is on camera, look for AI-generated faces and search your own customers for the same face under another name.
- 40+ document anti-fraud checks 40+ document checks
- Deepfake and face-swap detection Deepfake detection
- 1:N face search across your customers 1:N face search
- Included in KYC packages In KYC packages
Independently tested and certified


- ISO/IEC 27001:2022
- GDPR
- iBeta ready
16,000+ document types
from 195+ countries and territories
Customers verify with the ID they actually carry — in 90+ languages, including handwritten fields and MRZ.
document types
countries and territories
What Is Identity Fraud Prevention?
Identity fraud vs identity theft
Identity theft happens to a person: their details are stolen. Identity fraud is what the thief does next, opening an account, taking a loan or claiming a bonus in someone else's name. Identity fraud prevention software protects the business, and with it the people whose identities are misused.
Why fraud starts at onboarding
Once a fake account is open, it can receive scam money, take credit or claim offers. New account fraud detection at sign-up costs less than chasing the losses afterwards.
One session, four questions
Is the document genuine, does the face belong to it, is the person live, and have they signed up before under another name? Onboarding fraud prevention answers all four before the account opens.
Fraud prevention is not one check
No single check stops identity fraud. The document, the face, the liveness and your own customer base are checked together, in the same session.
Identity Fraud Schemes We Stop at Sign-Up
Six fraud schemes, one onboarding session.
Fraudsters reuse a handful of schemes to get past onboarding. Each one leaves a trace in the document, in the face or in your own customer base, and each is checked in the same KYC session.
Deepfakes and fake selfies
A deepfake video, a swapped face or an AI-generated selfie can look real to the person reviewing it. Every selfie is checked for generation and face swaps, and liveness blocks photos, screen replays, masks and deepfake videos shown to the camera.
Fake and altered ID documents
Photos of a document shown on another screen, photocopies, replaced photos, edited fields and data that contradicts the MRZ are flagged. Every document goes through 40+ anti-fraud checks across 16,000+ types from 195+ countries and territories.
Multi-accounting and duplicate sign-ups
Names, emails and documents are easy to change; a face is not. Each new selfie is searched against your existing customers, so one person cannot hold ten accounts under ten names.
Money mule accounts
Scam networks open accounts in bulk, with borrowed IDs or by people you have already rejected. Duplicate search, the selfie-to-document match and your own face blacklist stop that part at sign-up.
Bonus and promo abuse
Welcome bonuses, cashback and referral rewards attract people who sign up again and again, often with a relative's ID. Face search links every new account to the person who already claimed the offer.
Synthetic and stolen identities
Synthetic identity fraud mixes real and invented details into a person who does not exist; a stolen identity uses real details with the fraudster's face. Both have to pass the document checks, the face match and the liveness check.
How Fraud Detection Works in One KYC Session
Four layers of onboarding fraud detection, one session for the customer.
Checks run server-side; your team sees one result for the session.
Where Identity Fraud Enters Onboarding
Three steps of a sign-up, three different attacks, and a different set of checks on each.
A document that is not what it claims
The attacker brings a photo of a document from a screen, a photocopy, a document with a replaced photo or edited fields, or an invented document with data that does not add up.
- Screen and photocopy detection
- Photo substitution
- Edited fields
- Data consistency and MRZ
Every rejection has a reason
Each failed check returns its own code: display or screen, image edited, image printed, is deepfake, is not alive, black list. Your team sees which check rejected the session.
Fraud Rules You Control in the KYC Editor
Identity risk management without a release: the scenario decides what is checked and when a session is blocked.
Choose which checks run
Turn each document, selfie and liveness check on or off per scenario: strict for account opening, lighter for low-risk steps.
Set in the KYC editor, no release needed.
Block after failed attempts
Set how many critical errors end a session, so fraudsters cannot keep sending fake images until one of them passes.
Block mixed documents
Reject sessions where two different documents appear in one step, or where the same type of document comes back with different data.
Your own lists of faces
Keep returning customers in person lists and known fraudsters in a blacklist; a blacklist match blocks the session automatically.
One scenario, one audit trail for every decision.
What Onboarding Checks Prove, and What They Do Not
Deepfakes are checked, not assumed
Every selfie is checked for AI-generated faces, face swaps and fully generated images, and our liveness is resistant to video stream injection. We do not publish detection percentages.
Deepfake detectionDocuments are checked, not eyeballed
Screen photos, photocopies, replaced photos, edited fields and values that contradict the MRZ are flagged with a reason code, in the same check that reads the document.
Document fraud detectionWhat onboarding cannot see
A real person who opens an account in their own name and later hands it over passes every onboarding check. Catching that needs monitoring of how the account is used afterwards.
KYT for cryptoFraud prevention is a set of checks, not a guarantee
We stop the schemes that leave a trace in the document, in the face or in your own customer base, and we tell you which check failed. We do not promise that no fraudster ever gets through, and we do not publish detection percentages.
Identity Fraud Detection API
The same Web SDK and REST API you already use for KYC. There is no separate fraud API to integrate.
Ways to integrate
What every integration includes
- One result for the whole session
- A reason code for every failed check
- A callback with the final result
- A sandbox with test data
Cloud or on-premises
Your data can stay in your jurisdiction.
In the back office
- Session review
- Reason codes
- Person lists
- Blacklist
Identity Fraud Rules in Southeast Asia
Since 2023 regulators across the region have moved from asking banks to verify customers to asking them to keep scam and mule accounts out. Last reviewed: November 2026. This overview is for information only and is not legal advice.
Singapore
- Regulator
- MASIMDA
- At onboarding
- Where there is no face-to-face contact, customer due diligence must be at least as robust as it would be in person.
- Against scams and mule accounts
- A Shared Responsibility Framework for phishing scam losses has been in force since 16 December 2024.
- Regulator
- Bank Negara MalaysiaPayNet
- At onboarding
- The e-KYC policy document sets out what is required to onboard a customer electronically.
- Against scams and mule accounts
- The National Fraud Portal, run by PayNet, supports the National Scam Response Centre.
- Regulator
- OJK
- At onboarding
- Financial institutions must run an anti-fraud strategy with prevention and detection built in.
- Against scams and mule accounts
- The Indonesia Anti-Scam Centre has been operating since November 2024.
- Regulator
- BSP
- At onboarding
- Electronic KYC through a digital ID system is allowed, to the same standard as face-to-face identification.
- Against scams and mule accounts
- The Anti-Financial Account Scamming Act makes money muling a criminal offence.
- Regulator
- Bank of Thailand
- At onboarding
- A facial scan is required to open an account through a banking application.
- Against scams and mule accounts
- A facial scan is also required for transfers over 50,000 baht and for raising the daily limit above it.
- Regulator
- State Bank of Vietnam
- At onboarding
- Banks verify a customer's biometrics for electronic transactions.
- Against scams and mule accounts
- Biometric authentication is required for transfers over 10 million dong at once, or 20 million dong in a day.
Identity Fraud Prevention by Industry
Where each scheme hits hardest, and which checks close it at sign-up.
-
Banks and digital banks
Accounts opened in bulk and accounts opened with stolen IDs are the main risks; duplicate search and liveness close both at sign-up.
-
E-wallets and payments
High sign-up volumes attract bulk account creation and referral abuse; one face per account keeps them out.
-
Digital lending
Borrowers who apply under several names to take several loans at once are stopped by face search before the first disbursement.
-
Crypto exchanges
Deepfakes and borrowed IDs are used to open accounts for laundering; liveness, deepfake checks and a face blacklist stop repeat offenders.
-
Marketplaces
Fake sellers come back after a ban under a new name and a new document; the face does not change with the account.
-
Telecom
SIM and eSIM issue with a borrowed or altered ID is the usual entry point; the selfie-to-document match closes it at the counter.
Why teams choose NeuroVision Global Systems
Accuracy you can audit
Independently tested face matching, a confidence score on every result.
Accuracy you can audit
The face algorithm was tested by NIST and placed TOP-24, NIST FRVT 1:1 Verification. Every document and face result comes with a confidence score.
Live in under an hour
REST API, Web SDK, iOS and Android SDK, WebView; sandbox first.
Live in under an hour
Under an hour for REST API or Web SDK with sandbox keys; production in one business day. A 14-day trial on your real traffic comes with full support.
Global coverage
195+ countries and territories, 16,000+ document types, 90+ languages.
Global coverage
Customers verify with the ID they actually carry: 16,000+ document types from 195+ countries and territories, including handwritten fields and MRZ.
Anti-fraud built on real attacks
40+ document checks, 1:N face search, blacklist screening, 30 more layers.
Anti-fraud built on real attacks
Deepfake-resistant liveness, 1:N deduplication and blacklist screening on top of document checks — one person, one account.
Privacy by design
ISO/IEC 27001:2022, GDPR-ready; on-premises, cloud or hybrid.
Privacy by design
Run it in our cloud, in your data centres or hybrid — templates and personal data never leave your jurisdiction.
In-house engines only
OCR, face, liveness and anti-fraud are ours.
In-house engines only
One vendor covers KYC, KYB, AML and KYT on one API — one SLA, no third-party licences.
Identity Fraud Prevention Pricing
Fraud checks are part of the KYC packages: there is no separate anti-fraud fee.
Document package
From $0.25 per check
Document recognition with 40+ anti-fraud checks. Selfie, liveness, deepfake and face-swap checks and search by person lists come with the BASIC package.
BASIC package
Everything in DOCUMENT plus a selfie or liveness check, the face comparison, deepfake and face-swap checks and search by your person lists.
International screening
Sanctions, PEP and watchlist screening for the same customer in the same session, charged per check.
Our AI-Powered Identity Solutions
Everything you need to verify, authenticate, and protect your customers — all in one powerful platform.
A real person with a real document — verified in seconds. Document capture, face match, liveness, anti-fraud and AML.
Learn moreScreen every customer against sanctions, PEP and watchlists, including self-exclusion lists — inside the same verification flow.
Learn moreVerify the business behind the account: company registry data and ultimate beneficial owners (UBO), checked before onboarding.
Learn moreTransaction monitoring in real time: suspicious transfers are flagged for review, on the same API as KYC and AML.
Learn moreA real, present person — deepfake-resistant. Authenticate returning customers by face instead of passwords.
Learn morePayments confirmed with a face or an ID document — the identity your customer already verified, at checkout.
Learn moreTest Your Onboarding Against Fraud
Tell us which fraud you see most: deepfakes, fake IDs, duplicate accounts, mules or bonus abuse. We will set up a sandbox with the right checks and run it on test sessions with you.
Identity Fraud Prevention FAQ
Nine questions buyers ask before they connect fraud checks to their onboarding.
Didn’t find your answer?
Identity fraud prevention is stopping people from opening accounts with fake, stolen or borrowed identities. At onboarding it means checking that the document is genuine, the face matches it, the person is live, and they have not signed up before under another name.
Yes. The selfie is checked for AI-generated faces, face swaps and fully generated images, and liveness blocks deepfake videos shown to the camera, along with photos, screens and masks.
Read more about liveness detection and the attacks it stops.
Yes. Our liveness is resistant to video stream injection, when a prepared video is fed in place of the camera. We do not publish detection percentages.
Each new face is searched against your existing customers. If the same face appears under another name or document, the session is flagged and you see the earlier sessions with that face.
Read more about face verification and 1:N search across your customers.
They stop mule accounts opened in bulk, with borrowed IDs or by people on your blacklist. A real person who opens an account and later hands it over passes onboarding; catching that needs transaction monitoring.
Read more about transaction monitoring for crypto.
Invented documents fail the tampering and data-consistency checks, AI-generated faces are flagged, and the selfie still has to match the document photo and pass the liveness check.
Read more about document verification and the checks on every ID.
Yes. Add their face to your blacklist from the session view. A later attempt with that face is rejected automatically, whatever document or name is used.
The checks run while the customer moves through the flow, not as a separate waiting step. The customer photographs the document and takes a selfie, and the result comes back for the session as a whole.
Fraud checks are included in the KYC packages: document anti-fraud checks come with the DOCUMENT package, and the selfie, liveness and deepfake checks with BASIC.
See our pricing page for the current rates.


