Who Explains Privacy Policies to Me? Embodied and Textual LLM-Powered Privacy Assistants in Virtual Reality

2026-03-02

AI Summary

This study addresses the challenge of achieving meaningful informed consent in virtual reality (VR) environments, where systems collect extensive sensitive data yet privacy policies are often lengthy, complex, and decoupled from user interaction. To bridge this gap, the authors introduce a novel paradigm that embeds a large language model (LLM)-driven privacy assistant—delivered through both an embodied virtual avatar (via voice) and text-based chat—directly into a VR application store, providing just-in-time explanations at users’ decision points. A user study (N=21) demonstrates that both modalities foster deeper engagement with privacy information and more deliberate decision-making, with privacy scores primarily serving as veto criteria. While the text modality better supports reflection and review, the effectiveness of embodied interaction varies across individuals. This work offers a practical framework for operationalizing informed consent in immersive VR contexts.

Technology Category

• Humans and AI: Game Design — Virtual Humans, NPCs and Autonomous Characters
• Machine Learning: Privacy
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval

Application Category

• User Modeling, Personalization and Recommendation: User privacy protection in personalized systems
• Responsible Web: Consent frameworks and practices on the web
• Search and Retrieval-Augmented AI: Vertical and domain-specific search

Abstract

Virtual Reality (VR) systems collect fine-grained behavioral and biometric data, yet privacy policies are rarely read or understood due to their complex language, length, and poor integration into users' interaction workflows. To lower the barrier to informed consent at the point of choice, we explore a Large Language Model (LLM)-powered privacy assistant embedded into a VR app store to support privacy-aware app selection. The assistant is realized in two interaction modes: a text-based chat interface and an embodied virtual avatar providing spoken explanations. We report on an exploratory within-subjects study $(N = 21)$ in which participants browsed VR productivity applications under unassisted and assisted conditions. Our findings suggest that both interaction modes support more deliberate engagement with privacy information and decision-making, with privacy scores primarily functioning as a veto mechanism rather than a primary selection driver. The impact of embodied interaction varied between participants, while textual interaction supported reflective review.

Problem

• privacy policies
• Virtual Reality
• informed consent
• user comprehension
• privacy decision-making

Innovation

• LLM-powered privacy assistant
• embodied avatar
• virtual reality
• privacy policy explanation
• informed consent