FaceIn.id uses a zero-knowledge architecture where your biometric data never leaves your device. Our servers verify proofs, not biometrics. Even if we were breached, there's nothing to steal.
Your face or fingerprint is processed on your device. Only a mathematical proof — never your biometric data — is sent to our servers. No data to phish or hack.
Camera or fingerprint sensor captures biometric → Device's secure biometric hardware generates template → Template stored in Secure Enclave / StrongBox
Receives mathematical proof → Verifies proof validity → Issues authentication token → Never processes biometric data
Receives auth token → Verifies with FaceIn.id API → User is authenticated → No password involved at any step
Three independent security layers protect every authentication event. Each layer operates autonomously — compromise of one does not affect the others.
Phone theft is rising in cities worldwide — moped snatching, grab-and-run. FaceIn turns a stolen phone from an identity catastrophe into a hardware inconvenience.
A thief grabs your unlocked phone:
A thief grabs your unlocked phone:
The real-world difference
Urban phone theft targets unlocked phones specifically because a stolen unlocked device is a skeleton key to your entire digital life. FaceIn breaks that assumption. The phone is just glass and silicon without the owner's face.
FaceIn never sees your face. Your device performs the biometric check and returns a cryptographic attestation; FaceIn verifies that attestation. A spoof — photo, video, mask, or deepfake — that fails your device's check never produces a valid signature, so it never becomes a login.
Every sign-in uses a fresh, single-use challenge, so a captured or replayed signature is worthless. There is no face template or image on our servers to copy, and nothing about the login can be reused.
Industry-standard cryptographic primitives. No proprietary algorithms. No security through obscurity.
Symmetric encryption for data at rest. NIST-approved, used by the US government for classified data.
Used for: Encrypts vault entries, user preferences, and local biometric templates on device.
Elliptic curve digital signatures for authentication proofs. Same algorithm securing TLS certificates worldwide.
Used for: Signs cryptographic attestations and device verification challenges.
Elliptic curve Diffie-Hellman for secure key exchange. Generates ephemeral session keys.
Used for: Establishes encrypted channels between device and server.
HMAC-based key derivation function. Derives unique encryption keys from shared secrets.
Used for: Generates per-session encryption keys from device-server handshake.
FaceIn is designed for full compliance with global privacy and biometric data regulations from day one.
Illinois Biometric Information Privacy Act
FaceIn.id never collects, captures, stores, or transmits biometric identifiers to our servers. All biometric processing occurs exclusively on-device, meeting BIPA's strictest requirements by architectural design.
General Data Protection Regulation (EU)
Built with data minimization, purpose limitation, and privacy-by-design principles. Users have full control: access, portability, deletion, and the right to be forgotten — all self-service.
California Consumer Privacy Act
FaceIn.id provides all CCPA-mandated disclosures, opt-out mechanisms, and data deletion capabilities. We do not sell personal information.
Service Organization Control 2
Currently undergoing SOC 2 Type II certification for Security, Availability, and Confidentiality trust service criteria. Expected completion Q3 2026.
We're happy to walk through our architecture, share our security whitepaper, or answer any questions your security team has.
CONTACT SECURITY TEAM →