Text-Based Personas for Simulating User Privacy Decisions

TL;DR

Narriva generates human-readable synthetic privacy personas grounded in prior user decisions, improving privacy decision simulation accuracy and efficiency over baseline methods.

Contribution

It introduces a novel persona generation method based on actual privacy decisions, enhancing simulation accuracy and reducing prompt tokens compared to existing approaches.

Findings

Achieves up to 87% predictive accuracy in modeling privacy preferences.

Reduces prompt tokens by 80-95% compared to in-context learning.

Synthesized personas from one survey replicate behaviors of different studies.

Abstract

The ability to simulate human privacy decisions has significant implications for aligning autonomous agents with individual intent and conducting cost-effective, large-scale privacy-centric user studies. Prior approaches prompt Large Language Models (LLMs) with natural language user statements, data-sharing histories, or demographic attributes to simulate privacy decisions. These approaches, however, fail to balance individual-level accuracy, human auditability, token efficiency, and population-level representation. We present Narriva, an approach that generates text-based synthetic privacy personas to address these shortcomings. Narriva grounds persona generation in prior user privacy decisions, such as those from large-scale survey datasets, rather than purely relying on demographic stereotypes. It compresses this data into concise, human-readable summaries structured by established privacy theories. Through benchmarking across five diverse datasets, we analyze the characteristics of Narriva's synthetic personas in modeling both individual and population-level privacy preferences. We find that grounding personas in past privacy behaviors achieves up to 87% predictive accuracy, improving over a non-personalized LLM baseline by 6-17 percentage points across datasets, while yielding an 80-95% reduction in prompt tokens compared to in-context learning with raw examples. Finally, we demonstrate that personas synthesized from a single survey can reproduce the aggregate privacy behaviors and statistical distributions of entirely different studies.