CAREER: Moderation and Accountability on Private Communication Platforms

University Of North Carolina At Chapel Hill, Chapel Hill NC

Investigators

Abstract

Messaging applications that provide strong privacy for their users can have the unintended side effect of making it harder for users to report abuse. When the platform cannot see the contents of users' messages, or does not know which users are talking to each other, it can never know if a report of abusive content is real or not. This project's novelties are in new cryptographic approaches to design private messaging applications that allow users to verifiably report abusive content to the platform without compromising the privacy of unreported messages or conversations. The project's broader significance and importance is to enable new possibilities for moderation on private communication platforms. This project achieves multiple order of magnitude performance improvements over prior work by taking a co-design approach to the design of abuse reporting features for end-to-end encrypted or metadata-hiding messaging applications. It takes advantage of the existing structure of these systems to improve on both the functionality and performance possible via the more generic approaches of prior work. For example, instead of only assuming that an existing platform hides message metadata, the project designs solutions for metadata-hiding platforms that specifically rely on broad classes of techniques like secret sharing or mixnets. The project also develops new educational materials to allow both college undergraduates and middle/high school students to gain early exposure to security and privacy topics. The project's impact will be to advance the research and industry communities' ability to design and deploy new messaging systems that keep users safe and protect their privacy. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.