Phishing attacks represent one of the most pervasive threats to Internet security, affecting billions of users worldwide. Adversaries deploy increasingly sophisticated deceptive websites that impersonate legitimate services, including PayPal, Facebook, and major financial institutions, to harvest sensitive personal information such as credentials and social security numbers. According to the FBI’s latest report, phishing attacks resulted in over $10 billion in losses in a single year. Despite these staggering figures, the true impact of phishing is frequently underestimated, with many perceiving these attacks as merely compromising social media accounts. In reality, successful phishing campaigns can trigger substantial financial losses, political instability, and serious threats to national security.
Doowon Kim,
University of Tennessee, Knoxville
In this presentation, I will discuss our recent research that better understands phishing attacks through an attacker-centric lens, revealing the strategies and techniques adversaries employ to design and execute successful campaigns. Building on these insights, I will present our comprehensive evaluation of state-of-the-art anti-phishing systems (i.e., phishing detectors) using real-world phishing datasets. Our analysis exposes critical flaws and limitations in current defensive approaches that attackers routinely exploit. Finally, I will propose actionable design principles for developing more robust anti-phishing defenses capable of countering evolving adversarial tactics and closing the gaps in existing protection mechanisms.
Professor Doowon Kim is an Assistant Professor of Computer Science at the University of Tennessee, Knoxville. He received his Ph.D. in Computer Science from the University of Maryland, College Park in 2020, where he worked with Prof. Tudor Dumitras. His research interests span computer security (data-driven security and usable security), computer networks (Internet measurement), and Web security, with a particular focus on phishing attacks, recently. His work seeks to identify the root causes of security threats by understanding the actors involved (both adversaries and end-users) through data-driven and human-centered approaches. His research has been published in top-tier security conferences, including IEEE S&P, ACM CCS, USENIX Security, NDSS, ACM IMC, and theWebConf. He has received several prestigious awards, including the NSF CAREER Award (2025), the NSA Best Scientific Cybersecurity Paper Award (2017), and the Ann G. Wylie Dissertation Fellowship (2019).