Efficient Abuse Reporting for Private Messaging Systems

Abuse reporting systems like message franking allow end-to-end encrypted (E2EE) messaging platforms to verify user-generated abuse reports as part of their platform content moderation policies. Unfortunately, existing abuse reporting systems make several assumptions about the messaging platform architecture, or desired privacy guarantees, that limit what policy decisions are possible. Meta's lightweight message franking protocol used in Messenger, for example, assumes that the messaging platform itself undertakes all moderation responsibilities, and the platform must see all messaging metadata. Existing systems also do not provide mechanisms for a moderator to enforce their decisions. They instead stop at the point of convincing a moderator that reports are genuine. While partial or full solutions exist to address some of these limitations, they incur performance overheads that make deployment impractical for real-world systems. In my dissertation, I present three novel techniques to implement abuse reporting for E2EE messaging platforms. These offer practical performance for different deployment considerations, allowing for (a) abuse reporting for metadata-hiding messaging based on onion encryption, (b) moderation duties performed by third parties separate from the messaging platform, and (c) abuse reporting and moderation enforcement in platforms with third-party moderators.