> ## Documentation Index
> Fetch the complete documentation index at: https://docs.verisoul.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Security and Accuracy

> Understanding the security features and accuracy metrics of ID Check

## Security

<table className="w-full border-collapse">
  <thead>
    <tr>
      <th className="p-3 text-left bg-gray-100 border">Risk</th>
      <th className="p-3 text-left bg-gray-100 border">Security Functionality</th>
    </tr>
  </thead>

  <tbody>
    <tr>
      <td className="p-3 border">2D Images, Print Outs</td>
      <td className="p-3 border">3D FaceScan detects points in three-dimensions, meaning 2D images and prints are quickly dismissed</td>
    </tr>

    <tr>
      <td className="p-3 border">3D Masks, Ultra-realistic wax sculptures</td>

      <td className="p-3 border">
        A) 3D Liveness detection checks for liveness context throughout user's face over time, meaning static or partially static faces (partially altered with wax/mask) will be detected<br /><br />
        B) Active movement ensures user liveness; additionally, multiple user distances from camera generates high-fidelity liveness data
      </td>
    </tr>

    <tr>
      <td className="p-3 border">Video injection & DeepFakes</td>

      <td className="p-3 border">
        Technology detects when camera feed is being altered or user is trying to inject video<br /><br />
        Additionally, technology detects deepfake videos & images
      </td>
    </tr>

    <tr>
      <td className="p-3 border">FaceScan alteration</td>
      <td className="p-3 border">FaceScans are encrypted to prevent alteration on the client side</td>
    </tr>

    <tr>
      <td className="p-3 border">Client-side device risk</td>

      <td className="p-3 border">
        A) SDK checking for risk signals on the device that would indicate likely fraud<br /><br />
        B) Use obfuscation and checksums to ensure code base is not tampered with
      </td>
    </tr>
  </tbody>
</table>

## Accuracy

### Definitions

Imagine User A is already enrolled...

**False Acceptance Rate (FAR)** - this is the probability that a given user B can pretend to be User A

* This is the value that matters for authentication (getting into an existing account)

**False Rejection Rate (FRR)** - this is the probability that User A will be rejected when trying to authenticate again

* This is the value that matters for uniqueness (preventing an existing user from falsely creating a duplicate / new account)

### Accuracy:

* **FAR: 1 / 125,000,000 chance** (Apple's touch ID is 1/50K, and FaceID is 1/1M)
* **FRR: \< 3 / 100,000**
* **Works with beards, transparent glasses (not sunglasses), and makeup**
* **3-Dimensional** modeling based on facial features results in skin-tone agnostic accuracy