Talk Title:
Cyber-physical system security: Exploiting the physics of sensors to undermine AI-based decisions
Abstract:
Unprecedented advancements in sensing and AI allow systems to accurately measure their surroundings and take complex decisions without human supervision. From home smart devices to autonomous transportation, these cyber-physical systems are typically designed to withstand both traditional network and OS attacks. However, they are still prone to hardware vulnerabilities at the physical level which can heavily affect downstream AI-based decisions. This webinar will provide an overview of the relevant scope, vulnerabilities, attacker motivations, and challenges to build trustworthy intelligent systems.
Speaker Bio:
Dr. Sara Rampazzi is an Assistant Professor in the Department of Computer and Information Science and Engineering (CISE) at the University of Florida. Her research focuses on the security and privacy of cyber-physical systems, embedded systems design, modeling, and simulation techniques with applications in Healthcare, Autonomous Systems, and the Internet of Things.
Her research on injecting inaudible and invisible commands to smart home devices, phones, and tables LightCommands has been covered by several media outlets, including New York Times, CNN, and Wired. Her work on Lidar spoofing attacks on autonomous vehicles has been recognized by Forbes as the first practical attack against LiDAR-based perception of autonomous driving systems. Her research is funded by NSF, Facebook, and Toyota. More details can be found here - http://sararampazzi.com/
