Embedded Files
The conference proceedings of the Track 1 papers are available for downloading
From Learning to Action: AI Agents on the Cybersecurity Frontline
Jake Cho
Assistant Professor,Computer Science at Lewis University
Research Associate at Illinois Institute of Technology, Chicago, USA
Jake Cho is an Assistant Professor of Computer Science at Lewis University and a Research Associate at the Illinois Institute of Technology. He brings over 15 years of experience in system and network security, with a primary focus on machine learning based intrusion detection, embedded system security, and secure architecture design for mobile and IoT platforms. His earlier work includes leading security development for Samsung Galaxy smartphones and the Artik IoT platform, where he helped establish secure development practices and quality assurance processes integrated throughout the product lifecycle.
His research team focuses on advanced intrusion detection systems and intelligent threat response for both IT/IoT and industrial control system environments. Recently, his team has begun developing compact generative AI models purpose built for cybersecurity. These models are designed to adapt across a wide range of deployment targets, from resource constrained devices to large scale satellite networks, and across diverse infrastructures including general enterprise IT and critical industrial systems. His work aims to bridge the gap between AI innovation and real world cybersecurity defense.
Abstract of the Invited Talk
As cyber threats continue to evolve in scale and sophistication, organizations are increasingly relying on artificial intelligence to support critical security operations. At the center of this shift is a fundamental question: what roles should AI agents take on in cybersecurity, and how can we ensure they learn the right behaviors to fulfill those roles effectively? The challenge is not only technical but conceptual. It involves defining clear goals for AI agents, designing learning environments that reflect real-world threat dynamics, and ensuring that these agents can work alongside human experts with transparency and trust. In this talk, I will explore recent research on adaptive AI systems for cybersecurity, focusing on how to shape agent capabilities, evaluate performance in complex environments, and support collaborative decision-making between humans and machines. I will also introduce a new AI research center in the Midwest of the United States, which aims to advance this work through interdisciplinary collaboration and regional innovation.
Report abuse