SaTC: CORE: Small: Consistent and Private Group Communication

Texting and social media-based messaging applications have become nearly as common as face-to-face communications for conversation between individuals and groups. While it is known how to provide privacy for conversations between two individuals, there is a gap in extending these techniques to group conversations. This project is developing new communication techniques and open-source software for private communication among groups of users. These techniques and software enable new long-distance communication and collaboration in situations where privacy is important, which will benefit society by allowing groups to communicate without sharing their private data with advertisers and data brokers.

While several protocols are known that can provide strong privacy protections for two-party conversations, less is understood about the properties of private group conversations. In particular, neither deployed protocols such as Signal, nor academic protocols such as multiparty OTR and GOTR, can guarantee that participants in a group conversation maintain a consistent view of the set of messages, order of messages, or even the list of fellow participants in the conversation. Meanwhile, reliable broadcast protocols can provide some of these consistency properties but do not have the same privacy goals, carry a heavy bandwidth overhead, and are ill-suited for deployment in the modern network environment.

This project is developing new algorithms and software to close this gap between deployable private, two-party communication schemes and high-overhead, reliable group communication primitives. It has three specific aims:

1. By extending two-party privacy notions to a group context, and relaxing reliable broadcast requirements to a consistency formalism, the project is developing formal definitions, provably secure protocols, and software for private group messaging in realistic network settings.

2. In order to practically carry out private group conversations, messaging software must support several additional functionalities while preserving privacy.

Furthermore, both the core communication and these additional functions must

be usable, in that users can understand and act on the information about these protocols

provided by the implementations. The project is developing and implementing protocols for these tasks, while studying the usability of these protocols and implementations.

3. Beyond group messaging, the project is investigating notions of privacy and consistency in other group settings, such as voice and video communication, semi-public forums and closed mailing lists, and adapting the tools from the first two aims to address these settings. The project is expected to result in new efficient algorithms for private group communication and supporting tasks, training of undergraduate and graduate

students, and the broad dissemination of free software that promotes private group communication.

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.