Iasonas Polakis

Papers

Total Citations

130

H-Index

About

Iasonas Polakis is a leading researcher in security and privacy, with a focus on the adversarial arms race between automated attacks and defense mechanisms. His most influential work, "I am Robot: (Deep) Learning to Break Semantic Image CAPTCHAs" (2016, 130 citations), demonstrated how deep learning could defeat advanced image-based CAPTCHAs that were considered robust, exposing critical vulnerabilities in widely used security systems. This research reshaped the design of CAPTCHAs and highlighted the limitations of semantic challenges against modern AI. Polakis’s broader contributions span web security, authentication, and usable privacy, where he systematically uncovers how attackers exploit design flaws in human-interactive security measures. His work has been recognized with multiple best paper awards and has influenced both academic research and industry practices. By combining rigorous empirical analysis with a deep understanding of adversarial behavior, Polakis continues to advance the field, helping to build more resilient defenses against evolving threats. His research is essential reading for anyone studying the intersection of machine learning, human-computer interaction, and security.