Protecting User Data Through Privacy-Sensitive Robot Design

Published in HRI 25: ACM/IEEE International Conference on Human-Robot Interaction, 2025

Recommended citation: Sullivan, D. & Mutlu, B. 2025. "Protecting User Data Through Privacy-Sensitive Robot Design." In Companion of the 2025 ACM/IEEE International Conference on Human-Robot Interaction (HRI 25). Association for Computing Machinery, New York, NY, USA.

Download Paper Here

Abstract: While robots possess many capabilities that may positively influence human lives, their autonomous navigation and sensing capabilities pose threats to user privacy. These threats may be addressed at three key phases: data collection, data retention, and data exposure. In this work, we discuss our prior, current, and proposed robot design efforts to reduce privacy violations during human-robot interaction (HRI). At the data collection phase, we are currently exploring designs that enable robots to inhibit data collection by blocking their own sensors. At the data retention phase, we propose the exploration of privacy preferences to inform designs that grant users greater control over retained data. Finally, in the data exposure phase, we discuss our prior works developing a privacy controller for appropriate data exposure and generating task-planning strategies to limit unintentional data exposure. Through this work, we hope to protect user data and reduce the likelihood of harm to users.