Uploaded June 2026 | Updated September 2026, 3 weeks ago
Vision: Human-as-the-Unit Privacy Management with AI Agents
Eryue Xu, Northeastern University and University of Illinois Urbana–Champaign
Managing one's digital footprint is overwhelming, as it spans multiple platforms and involves countless context-dependent decisions. This talk presents research from the CHI 2026 paper by Eryue Xu and Tianshi Li, which explores how emerging agentic AI systems might support more comprehensive privacy management. We adopted a "human-as-the-unit" perspective and investigated users' cross-context privacy challenges through semi-structured interviews. Results reveal that people rely on ad hoc manual strategies while lacking comprehensive privacy controls, highlighting nine privacy-management challenges across applications, temporal contexts, and relationships. To explore solutions, we generated nine AI agent concepts and evaluated them via a speed-dating survey with 116 US participants. The highest-ranked designs focused on post-sharing privacy management, where AI agents help users detect, review, and remediate previously shared information. Participants expressed strong interest in automated assistance and often reported greater confidence in AI-supported privacy management than in their own manual efforts. Our findings highlight a promising design space where users see AI agents bridging the fragments in privacy management.
View the full PEPR '26 program at usenix.org/conference/pepr26/program
Vision: Human-as-the-Unit Privacy Management with AI Agents
Eryue Xu, Northeastern University and University of Illinois Urbana–Champaign
Managing one's digital footprint is overwhelming, as it spans multiple platforms and involves countless context-dependent decisions. This talk presents research from the CHI 2026 paper by Eryue Xu and Tianshi Li, which explores how emerging agentic AI systems might support more comprehensive privacy management. We adopted a "human-as-the-unit" perspective and investigated users' cross-context privacy challenges through semi-structured interviews. Results reveal that people rely on ad hoc manual strategies while lacking comprehensive privacy controls, highlighting nine privacy-management challenges across applications, temporal contexts, and relationships. To explore solutions, we generated nine AI agent concepts and evaluated them via a speed-dating survey with 116 US participants. The highest-ranked designs focused on post-sharing privacy management, where AI agents help users detect, review, and remediate previously shared information. Participants expressed strong interest in automated assistance and often reported greater confidence in AI-supported privacy management than in their own manual efforts. Our findings highlight a promising design space where users see AI agents bridging the fragments in privacy management.
View the full PEPR '26 program at usenix.org/conference/pepr26/program
