PersonaFingerprint: Measuring Persona Inference on Modern Websites with LLM-Driven Browsing

TL;DR

This paper demonstrates that encrypted traffic metadata can reveal user personas and browsing behaviors on modern websites, using an LLM-driven framework to quantify and improve inference accuracy.

Contribution

It introduces a novel multi-agent browsing framework and formalizes persona fingerprinting, revealing significant privacy risks in modern web traffic analysis.

Findings

Persona inference accuracy reaches 84% on mixed-site traffic.

A lightweight multi-task model boosts persona accuracy to 80%.

Encrypted traffic metadata leaks both site and user persona information.

Abstract

Website Fingerprinting (WFP) has traditionally focused on inferring which website a user visits from encrypted traffic metadata such as packet sizes and timing. In this paper, we identify and quantify a new privacy risk in modern web settings: an adversary can infer a user's persona using only packet-length and inter-arrival-time sequences. To study this risk at scale, we build an LLM-driven multi-agent browsing framework that enforces controllable persona constraints while a computer-use agent interacts with real websites and collects corresponding encrypted traffic traces. We formalize persona fingerprinting under both closed-set and open-world settings and further evaluate whether persona information is already embedded in representations learned by existing WFP models and can be amplified at low cost. Across 10 modern websites and 15 personas (plus an open-world class), persona inference achieves about 84% accuracy on mixed-site traffic; moreover, a lightweight multi-task objective can boost persona accuracy to around 80% while retaining strong site classification performance (about 93% baseline). Our results show that, on modern websites, encrypted traffic metadata can leak not only which site a user visits, but also how they browse and who is browsing.