Rescriber: Smaller-LLM-Powered User-Led Data Minimization for LLM-Based Chatbots

TL;DR

Rescriber is a browser extension that enables users to detect and sanitize personal information in prompts for LLM-based chatbots, enhancing privacy control and reducing sensitive data disclosure.

Contribution

This work introduces Rescriber, a novel user-led data minimization tool powered by smaller LLMs, allowing personalized privacy management in conversational agents.

Findings

Rescriber effectively reduces unnecessary personal data disclosure.

User trust correlates with detection and sanitization quality.

Smaller LLMs can power on-device privacy controls comparable to larger models.

Abstract

The proliferation of LLM-based conversational agents has resulted in excessive disclosure of identifiable or sensitive information. However, existing technologies fail to offer perceptible control or account for users' personal preferences about privacy-utility tradeoffs due to the lack of user involvement. To bridge this gap, we designed, built, and evaluated Rescriber, a browser extension that supports user-led data minimization in LLM-based conversational agents by helping users detect and sanitize personal information in their prompts. Our studies (N=12) showed that Rescriber helped users reduce unnecessary disclosure and addressed their privacy concerns. Users' subjective perceptions of the system powered by Llama3-8B were on par with that by GPT-4o. The comprehensiveness and consistency of the detection and sanitization emerge as essential factors that affect users' trust and perceived protection. Our findings confirm the viability of smaller-LLM-powered, user-facing, on-device privacy controls, presenting a promising approach to address the privacy and trust challenges of AI.