Privacy in Action: Towards Realistic Privacy Mitigation and Evaluation for LLM-Powered Agents

Shouju Wang, Fenglin Yu, Xirui Liu, Xiaoting Qin, Jue Zhang, Qingwei Lin, Dongmei Zhang, Saravan Rajmohan


Abstract
The increasing autonomy of LLM agents in handling sensitive communications, accelerated by Model Context Protocol (MCP) and Agent-to-Agent (A2A) frameworks, creates urgent privacy challenges. While recent work reveals significant gaps between LLMs’ privacy Q&A performance and their agent behavior, existing benchmarks remain limited to static, simplified scenarios. We present PrivacyChecker, a model-agnostic, contextual integrity based mitigation approach that effectively reduces privacy leakage from 36.08% to 7.30% on DeepSeek-R1 and from 33.06% to 8.32% on GPT-4o, all while preserving task helpfulness. We also introduce PrivacyLens-Live, transforming static benchmarks into dynamic MCP and A2A environments that reveal substantially higher privacy risks in practical. Our modular mitigation approach integrates seamlessly into agent protocols through three deployment strategies, providing practical privacy protection for the emerging agentic ecosystem. Our data and code will be made available at https://aka.ms/privacy_in_action.
Anthology ID:
2025.findings-emnlp.925
Volume:
Findings of the Association for Computational Linguistics: EMNLP 2025
Month:
November
Year:
2025
Address:
Suzhou, China
Editors:
Christos Christodoulopoulos, Tanmoy Chakraborty, Carolyn Rose, Violet Peng
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
17055–17074
Language:
URL:
https://aclanthology.org/2025.findings-emnlp.925/
DOI:
Bibkey:
Cite (ACL):
Shouju Wang, Fenglin Yu, Xirui Liu, Xiaoting Qin, Jue Zhang, Qingwei Lin, Dongmei Zhang, and Saravan Rajmohan. 2025. Privacy in Action: Towards Realistic Privacy Mitigation and Evaluation for LLM-Powered Agents. In Findings of the Association for Computational Linguistics: EMNLP 2025, pages 17055–17074, Suzhou, China. Association for Computational Linguistics.
Cite (Informal):
Privacy in Action: Towards Realistic Privacy Mitigation and Evaluation for LLM-Powered Agents (Wang et al., Findings 2025)
Copy Citation:
PDF:
https://aclanthology.org/2025.findings-emnlp.925.pdf
Checklist:
 2025.findings-emnlp.925.checklist.pdf