Biometrics have a problem passwords do not: you cannot reset your face. If a service stores your fingerprint or facial template and is breached, that biometric is compromised forever — there is no rotating it. A stored template is therefore a permanent honeypot, and the security question is how to authenticate someone by biometric without keeping the biometric itself in a form an attacker can use.
Visa filed squarely in this space and landed two related grants in 2023. US11546164B2, issued January 3, 2023, and US11831780B2, issued November 28, 2023, share the title “Verification of biometric templates for privacy preserving authentication” and the same inventor team — Badrinarayanan, Rindal, Mukherjee — names associated with applied secure-computation research.
The CPC tags — H04L 9/3231 for biometric-based authentication alongside H04L 9/0866 and H04L 9/3271 — point at the cryptographic core. The idea, reading the claims, is to verify that a presented biometric matches an enrolled one without the verifier ever holding the enrolled template in usable plaintext. The match happens over protected representations, so a breach yields nothing reusable.
Two grants with the same title and team, ten months apart, is a portfolio pattern, not a single filing. The earlier grant (B2 with a pre-grant publication) and the later continuation suggest Visa was building layered coverage around one technique — the kind of deliberate, defensive fencing a payments network does around a capability it expects to deploy at scale.
Per the desk's discipline: both are issued grants (B2), not applications, and they claim verification methods, not a shipped Visa product. The commercial logic is obvious for a payments network moving toward biometric authentication at the point of sale — you cannot afford to be the company that leaked a billion fingerprints.
For the IP reader, biometric-template protection is a quietly important corner of the privacy-preserving-computation map. It applies secure-computation and cryptographic-matching techniques to a consumer-scale problem, and when a network the size of Visa fences it with multiple grants, that is a signal the technology is heading into production.