Synopsis: Secure and private trend inference from encrypted semantic embeddings

2025-05-29

AI Summary

This work addresses the challenge of privacy-preserving trend analysis on end-to-end encrypted (E2EE) messaging platforms (e.g., WhatsApp), enabling trusted entities—such as journalists—to conduct exploratory or targeted analyses of misinformation or political content propagation without accessing raw messages. We propose a novel hybrid framework integrating local and centralized differential privacy with malicious-secure multi-party computation (MPC), enforcing a single, cryptographically guaranteed legitimate query path. Fine-grained trend aggregation is performed directly in the encrypted domain using voluntarily contributed 500-dimensional semantic embeddings. Evaluated on a dataset of 34,024 Hindi WhatsApp messages, our system achieves >94% analytical accuracy with ~30-second query latency, demonstrating strong privacy guarantees (formal DP bounds), practical efficiency, and flexible analytical expressiveness.

Technology Category

• Machine Learning: Privacy
• Natural Language Processing: Ethics — Bias, Fairness, Transparency & Privacy
• Data Mining & Knowledge Management: Representing, Reasoning, and Using Provenance, Trust

Application Category

• Security and Privacy: Data transparency and provenance
• User Modeling, Personalization and Recommendation: User privacy protection in personalized systems
• Responsible Web: Data and user privacy-enhancing technologies for the Web

Abstract

WhatsApp and many other commonly used communication platforms guarantee end-to-end encryption (E2EE), which requires that service providers lack the cryptographic keys to read communications on their own platforms. WhatsApp's privacy-preserving design makes it difficult to study important phenomena like the spread of misinformation or political messaging, as users have a clear expectation and desire for privacy and little incentive to forfeit that privacy in the process of handing over raw data to researchers, journalists, or other parties. We introduce Synopsis, a secure architecture for analyzing messaging trends in consensually-donated E2EE messages using message embeddings. Since the goal of this system is investigative journalism workflows, Synopsis must facilitate both exploratory and targeted analyses -- a challenge for systems using differential privacy (DP), and, for different reasons, a challenge for private computation approaches based on cryptography. To meet these challenges, we combine techniques from the local and central DP models and wrap the system in malicious-secure multi-party computation to ensure the DP query architecture is the only way to access messages, preventing any party from directly viewing stored message embeddings. Evaluations on a dataset of Hindi-language WhatsApp messages (34,024 messages represented as 500-dimensional embeddings) demonstrate the efficiency and accuracy of our approach. Queries on this data run in about 30 seconds, and the accuracy of the fine-grained interface exceeds 94% on benchmark tasks.

Problem

• Secure trend analysis from encrypted messaging platforms
• Privacy-preserving inference without raw data access
• Balancing exploratory and targeted analyses with differential privacy

Innovation

• Combines local and central differential privacy models
• Uses malicious-secure multi-party computation
• Analyzes encrypted messages via semantic embeddings