Initial Proposal Abstract: This project aims to design, prototype, evaluate, and potentially deploy an AI-enabled voice agent to assist patients (especially older adults) to better recognize phishing messages and reduce cybersecurity risks during patient outreach and communications.

Timely patient outreach and communication are critical to the well-being of patients and improve the efficiency of healthcare workers. Without security measures, such communication channels are vulnerable to phishing attacks (where malicious hackers trick the victims into revealing sensitive information). To help patients better recognize phishing messages and verify message authenticity, we propose to design and develop AI-enabled voice agents. The system includes (1) a security-word-based message verification scheme to verify message authenticity, and (2) an AI agent powered by Large Language Models (LLMs) to help patients with little to no technical background to recognize phishing attempts (based on signals including the security words) and take safe actions.

To evaluate the effectiveness and usability of the system, we plan to recruit 52 participants (including older adults) to interact with the voice agent and perform phishing detection tasks. We will collect both quantitative and qualitative data to assess and further improve the system. We will work with OSF Healthcare to explore deployment opportunities.

Outcomes:

• 
  Publication: Proceedings from SOUPS ’25 – Can You Walk Me Through It? Explainable SMS Phishing Detection using LLM-based AgentsJune 12, 2025
  Authors: Yizhu Wang, Haoyu Zhai, Chenkai Wang, Qingying Hao, Nick A Cohen, Roopa Foulger, Jonathan A Handler, and Gang Wang. Abstract SMS phishing poses a significant threat to users, especially older adults. Existing defenses mainly focus on phishing detection, but often cannot explain why the SMS is malicious to lay users. In… Read more: Publication: Proceedings from SOUPS ’25 – Can You Walk Me Through It? Explainable SMS Phishing Detection using LLM-based Agents
• 
  Poster Presentation: a2 National Symposium 2025April 4, 2025
  Title: Protecting Patients against Phishing Attacks using AI-enabled Agents Authors: Yizhu Wang, Haoyu Zhai, Chenkai Wang, Qingying Hao, Nick Cohen, Roopa Foulger, Jon Handler, Gang Wang
• 
  Oral Presentation: Mobile Computing Seminar at Columbia University (Virtual)December 29, 2024
  Dr. Gang Wang gave a virtual talk at Columbia University (mobile computing seminar) which highlighted this project and the prototype design. Title: Defending against Spoofing Attacks in Mobile Networks and Applications Presenter: Gang Wang (PI)