Privacy engineering and unlinkable credentials
All AmbiSecure engineering content tagged Privacy, newest first. Newer entries reflect current thinking; earlier entries come from the engineering archive.
Privacy in identity engineering is about proving what is necessary and revealing nothing more. Techniques such as match-on-device biometrics, pseudonymous certificates, selective disclosure, and origin-scoped credentials let a system authenticate a user without building a central pool of linkable personal data.
This category covers privacy-preserving design for credentials and connected systems: how biometric templates stay on-device, how V2X pseudonymous certificates prevent vehicle tracking, and how FIDO's per-origin credentials avoid cross-site correlation. The guiding principle is data minimisation enforced by architecture, not policy.
Articles tagged Privacy
Frequently asked questions
How do FIDO credentials protect privacy?
FIDO generates a distinct key pair per relying-party origin, so credentials cannot be used to correlate a user across sites, and no shared identifier is handed to every service.
What are pseudonymous certificates?
In V2X (vehicle-to-everything) systems, vehicles use short-lived pseudonymous certificates so messages are authenticated without exposing a persistent identifier that would let observers track a specific vehicle.
Why keep biometric templates on the device?
Match-on-device keeps the template in a secure element and never transmits it, so there is no central biometric database to breach and no raw biometric to leak.
Can two services correlate a user through their FIDO credentials?
No. A distinct key pair is generated per relying party, so there is no shared identifier for services to compare.
Why do V2X systems need pseudonymity at all?
Because a vehicle broadcasts safety messages continuously. A static identity would let any passive listener reconstruct its movements, so certificates rotate to limit how long a vehicle can be followed.