Daniel Zappala
I lead the Usable Security and Privacy Lab, where our goal is to help people protect their security and privacy online. Our research is human-centered, meaning we collaborate with people to understand and design for their needs.
We have a variety of projects underway.
Studying how people apply mental models and threat models to understand and reason about new authentication technologies.
Comparing the usability and adoption of Passkeys authentication with security tokens and passwords.
Identifying how usable security and privacy researchers conduct human-centric threat modeling.
Understanding how immigrant generations teach each other online security and privacy practices.
Led by Warda Usman
Secure email systems that use end-to-end encryption are the best method we have for ensuring user privacy and security in email communication. Despite its clear advantages, the adoption of secure email remains low, with previous studies suggesting mainly that secure email is too complex or inconvenient to use. However, the perspectives of those who have, in fact, chosen to use an encrypted email system are largely overlooked.
To understand these perspectives, we conducted a semi-structured interview study that aims to provide a comprehensive understanding of the mindsets underlying adoption and use of secure email services. Our participants come from a variety of countries and vary in the amount of time they have been using secure email, how often they use it, and whether they use it as their primary account.
Our results uncover that a defining reason for adopting a secure email system is to avoid surveillance from big tech companies. However, regardless of the complexity and accuracy of a person’s mental model, our participants rarely send and receive encrypted emails, thus not making full use of the privacy they could obtain. These findings indicate that secure email systems could potentially find greater adoption by appealing to their privacy advantages, but privacy gains will be limited until a critical mass are able to join these systems and easily send encrypted emails to each other. For details, read our SOUPS 2023 paper, Distrust of big tech and a desire for privacy: Understanding the motivations of people who have voluntarily adopted secure email
Led by James Conners
Passwords have numerous drawbacks, and as a result many systems have been designed to replace them. Password replacements have generally failed to dislodge passwords due to the complexity of balancing usability, deployability, and security. However, despite this lack of success, recent advances with password managers and FIDO2 afford new opportunities to explore system design for password replacements.
In this work, we explore the feasibility of a system for user authentication based on certificates. Rather than developing new cryptography, we develop a new system, called Let’s Authenticate, which combines elements of password managers, FIDO2, and certificates. Our design incorporates feedback from a survey of 397 participants to understand their preferences for system features. Let’s Authenticate issues privacy-preserving certificates to users, automatically manages their credentials, and eliminates trust in third parties.
We provide a detailed security and privacy analysis, an overhead analysis, and a systematic comparison of the system to a variety of alternatives using a well-known framework. We discuss how Let’s Authenticate compares to other systems, lessons learned from our design, and issues related to centralized management of authentication data. For details, read our 2022 NDSS paper, Let’s Authenticate: Automated Certificates for User Authentication.
This work was a forerunner of the FIDO Alliance PassKeys system. Both systems use cryptographic credentials that are stored in the cloud, provide privacy to users, and ensure easy account recovery.
Led by Tarun Yadav
Popular instant messaging applications such as WhatsApp and Signal provide end-to-end encryption for billions of users. These applications often rely on a centralized, application-specific server to distribute public keys and relay encrypted messages between the users. As a result, they prevent passive attacks but are vulnerable to some active attacks. A malicious or hacked server can distribute fake keys to users to perform man-in-the-middle or impersonation attacks. While typical secure messaging applications provide a manual method for users to detect these attacks, this burdens users, and studies show it is ineffective in practice.
This paper presents KTACA, a completely automated approach for key verification that is oblivious to users and easy to deploy. We motivate KTACA by designing two approaches to automatic key verification. One approach uses client auditing (KTCA) and the second uses anonymous key monitoring (AKM). Both have relatively inferior security properties, leading to KTACA, which combines these approaches to provide the best of both worlds.
We provide a security analysis of each defense, identifying which attacks they can automatically detect. We implement the active attacks to demonstrate they are possible, and we also create a prototype implementation of all the defenses to measure their performance and confirm their feasibility. Finally, we discuss the strengths and weaknesses of each defense, the load they impose on clients and service providers, and their deployment considerations.
For details, see our 2022 CCS paper, Automatic Detection of Fake Key Attacks in Secure Messaging, which was a collaboration among researchers at Brigham Young University, Max Planck Institute for Informatics, and University of Connecticut.