Secure but Usable? Evaluating Homomorphic Encryption in Biometric Authentication Systems

This thesis explores whether homomorphic encryption can be used in biometric authentication systems in a way that is not only secure but also usable in practice. By building and evaluating a prototype that performs biometric matching on encrypted data, the project will investigate how the latency and interaction delays introduced by homomorphic encryption affect user experience, trust, and acceptance. A particular focus will be on how waiting times are perceived, how system feedback should be designed, and where the usability limits of privacy-preserving authentication lie, with the goal of deriving concrete design recommendations for making encrypted biometric systems both privacy-preserving and usable for real users.