TABLE OF CONTENTS
Overview
This guide explains the checks used in our electronic identity verification (EIV) process. These checks confirm that an individual and their ID document are genuine and haven’t been tampered with or fraudulently used.
The process includes:
Anti-tampering checks: Verifying that an ID document is genuine and unaltered.
Anti-fraud checks: Identifying potential digital or behavioural signs of fraud.
Biometric checks: Ensuring the person presenting the ID is its true owner.
1. Anti-tampering:
Anti-tampering checks confirm whether an identity document (such as a passport, driver’s licence, or national ID) is authentic and unaltered. These checks review:
Data accuracy (the information on the document makes sense and matches other records)
Visual authenticity (the image looks real, not edited or fake)
Database records (the document hasn’t been listed as lost, stolen, or compromised)
The system also compares details taken from the document with the information the user provided elsewhere in the verification process. Any mismatch may indicate potential tampering or fraud.
Most checks are fully automated and completed within seconds. If a document image is unclear or shows signs of advanced counterfeiting, our trusted verification partner’s manual review team will review it in detail to confirm authenticity.
Possible results:
| Result | What it means |
|---|---|
| CLEAR | All checks passed. The document appears genuine. |
| CAUTION | Some checks failed, but there’s no clear sign of fraud (for example, a name mismatch). |
| SUSPECTED | The document may be fraudulent. |
| REJECTED | The document couldn’t be processed (for example, poor image quality or unsupported document type). |
Results are supported by Reasons that explain which parts of the verification failed or need review.
Example of a REJECTED result and the associated reasons:
Breakdown of Results and Explanations
1. REJECTED
| Reason | Sub-Reason | What It Means |
|---|---|---|
| Image Quality | Poor or unreadable image | The photo is too blurry, cropped, or dark to read. |
| Image Format | Black and white image | Security features can’t be verified without colour. |
| Document Type | Unsupported or sanctioned country | The document type isn’t supported or comes from a restricted country. |
| Age Validation | Below minimum accepted age | The date of birth shows the applicant is under the account’s minimum age (default 16). |
| Repeat Attempts | Same document reused | The same ID has been submitted multiple times suspiciously. |
2. SUSPECTED
| Reason | Sub-Reason | What It Means |
|---|---|---|
| Data Consistency | Mismatch between document fields | Details (like name, date of birth, gender) don’t match across the document or MRZ. |
| Visual Authenticity | Missing or altered face image | The system can’t detect a valid face photo on the ID. |
| ODP (Original Document Present) | Photo of a photo or screen | The image appears to be taken from another screen or printed copy. |
| Fonts | Unexpected font style | Fonts don’t match those used in genuine government-issued IDs. |
| Picture Integrity | Edited or replaced image | The ID photo shows signs of tampering or insertion. |
| Digital Tampering | Security features missing or altered | Holograms or watermarks appear edited or absent. |
| Template Mismatch | Layout differs from known template | The structure doesn’t match the expected format for that country’s ID. |
| Compromised Document | Appears in compromised ID databases | The ID number has been reported lost or stolen. |
3. CAUTION
| Reason | Sub-Reason | What It Means |
|---|---|---|
| Image Integrity | Minor image issues | The image is usable but has watermarks, glare, or cropped edges. |
| Document Expiration | Expired ID | The document’s expiry date has passed. |
| Data Comparison | Mismatch with submitted data | The document’s data doesn’t match information provided elsewhere (e.g., form input). |
| Data Validation | Invalid format | The MRZ or document number format doesn’t match official standards. |
2. Anti-Fraud
Anti-fraud checks look for signs of digital or behavioural fraud beyond the document itself. These checks analyse data such as:
Device information (hardware, software, camera feed, and metadata)
Document details (numbers, type, and structure)
IP address and location (including VPN or proxy use)
These signals help detect fraudulent behaviour without adding friction for genuine users.
What’s Assessed
| Category | Purpose |
|---|---|
| Application authenticity | Identifies whether someone is using stolen or fake login credentials. |
| Device integrity and reputation | Detects suspicious device activity, such as emulators or randomised device data. |
| Police record | Checks if the ID document was reported as lost, stolen, or compromised. |
| IP and geolocation | Compares the IP location to the country on the ID to flag possible risk if they don’t match. |
Device Intelligence Check Results
| Result | Meaning |
|---|---|
| CLEAR | The applicant used a valid device with no suspicious behaviour. |
| CONSIDER | The applicant used an invalid device or there are signs of suspicious activity. |
Device Intelligence: Consider Reasons
| Reason | Explanation |
|---|---|
| Application authenticity | The system detected fake or stolen network data. |
| Device integrity | The device showed unusual patterns, such as: <ul><li>Randomised or false device data</li><li>Use of an emulator</li><li>Multiple devices used for one application</li><li>Suspicious IP traffic</li></ul> |
Police Record Check Results
| Result | Meaning |
|---|---|
| CLEAR | The document hasn’t been reported to the UK Metropolitan Police. |
| CONSIDER | The document matches one reported as lost, stolen, or compromised. |
3. Biometrics
Biometric checks verify that the person completing the identity verification is the genuine document holder. The system compares the face on the user’s live photo or motion capture video to the face on their ID document.
If the document has two sides, both are checked for a matching face. These checks confirm that the document owner — not someone else — is the person being verified.
For high-risk cases, Facial Similarity Motion adds a “liveness” step, asking users to turn their head to confirm that the image is real and not a photo or video spoof.
Biometric Motion Check Results
| Result | Meaning |
|---|---|
| CLEAR | The person’s face matches their ID, and the motion appears live and genuine. |
| CONSIDER | There are signs that the motion capture or face match may not be genuine. |
Biometric Motion Consider Reasons
| Breakdown | Sub-Breakdown | Description |
|---|---|---|
| Visual authenticity | — | Confirms that the motion capture shows a real person, not a spoof or static image. |
| Spoofing detection | — | A score between 0 and 1 shows how real the motion capture appears (1 = genuine). |
| Liveness detected | — | Confirms that the user completed the required head movement pattern. |
| Face comparison | Face match | Compares the face on the ID with the live capture. The score ranges from 0 (no match) to 1 (perfect match). |
| Image integrity | Face detected | Confirms that a clear face was found in both the document image and live capture. |
| Source integrity | — | Checks whether the live photo came from a trusted source and wasn’t digitally tampered with or taken through a fake webcam. Also checks if the document’s issuing country is under sanctions. |
Summary
These checks work together to verify both the document and the person presenting it.
Anti-tampering ensures the ID document is genuine.
Anti-fraud identifies unusual or risky digital behaviour.
Biometrics confirm that the document belongs to the person completing verification.
Together, these steps create a secure and reliable identity verification process.