On October 29, 2026, the Faculty of Data Science at City University of Macau organized an academic seminar titled “Assessing and Protecting Privacy with AI Agents”. The seminar focused on the application of AI agents in privacy assessment and protection, with particular attention to the privacy risks that may arise during information reasoning, tool use, and sensitive information identification, as well as how agent-based technologies can be used to support privacy protection. The seminar aimed to help faculty members and students gain a deeper understanding of the privacy and security challenges emerging alongside the rapid development of artificial intelligence.
Prior to the lecture, Assistant Professor Huajie Chen from the Faculty of Data Science introduced the guest speaker to the audience and provided a brief overview of Dr. Yixin Wu’s academic background and research interests. The lecture was delivered by Dr. Yixin Wu, Assistant Professor in the AI and IoT Thrust at The Hong Kong University of Science and Technology (Guangzhou). Dr. Wu also serves as a cross-campus Faculty Affiliate of The Hong Kong University of Science and Technology. Her research focuses primarily on AI security, privacy and transparency, as well as AI agents for security and privacy applications. Her research has been published in leading international conferences and venues, including USENIX Security, ACM CCS and ECCV.
During the seminar, Prof. Wu began by discussing emerging privacy issues arising from the rapid development of AI agents. She introduced how agents can combine visual reasoning, external tools, and multi-step information analysis to infer potentially sensitive information from public or limited data, and further examined the privacy leakage risks that may result. Drawing on her related research, she also presented applications of AI agents in privacy risk identification, privacy assessment, and protection, giving participants a more direct and in-depth understanding of the relationship between increasingly capable AI agents and privacy protection.
During the discussion session, faculty members and students exchanged views with Prof. Wu on topics including privacy risks associated with AI agents, assessment methods, and future research directions. Prof. Wu provided detailed responses based on her own research experience, further broadening participants’ understanding of the relevant research questions.
The seminar provided a valuable platform for faculty members and students to learn about frontier research in AI agents, AI security, and privacy protection. It also helped broaden academic perspectives in these areas and deepen understanding of the security and privacy challenges brought about by advances in artificial intelligence. The Faculty of Data Science will continue to organize academic seminars and exchange activities on a wide range of topics, providing more opportunities for faculty members and students to engage with cutting-edge research, develop new research ideas, and promote academic exchange.
