Keynote Speakers
Prof. Bo Luo
H.J. and Joan O. Wertz Endowed Professor, University of Kansas
Title: Securing AI, Abusing AI, Trusting AI: Three Views of Security and Privacy in Machine Learning Models and Applications
Abstract: Machine learning, especially generative AI, has become both a new target and a new tool in cybersecurity: machine learning models can be poisoned, backdoored, stolen, or manipulated; generative AI can be abused to produce deceptive content, automate misuse, and amplify social and technical threats; and AI-driven systems raise new questions about robustness, privacy, authenticity, and trust. In this talk, I will discuss the evolving relationship between cybersecurity and AI through three perspectives: securing AI, abusing AI, and trusting AI. Through some examples from our group's recent work on adversarial machine learning, AI-generated content detection, and trustworthy generative AI, I would like to highlight emerging security and privacy challenges in machine learning models and applications, and to identify research opportunities at the intersection of AI and cybersecurity.
Bio: Bo Luo is the H.J. and Joan O. Wertz Endowed Professor with the EECS department at the University of Kansas. He is also the director of the Center for High Assurance and Secure Systems (HASS) at KU's Institute of Information Sciences (I2S). He received Ph.D. degree from The Pennsylvania State University in 2008, M.Phil degree from the Chinese University of Hong Kong in 2003, and B.E. from University of Sciences and Technology of China in 2001. His recent works mostly lie in the intersection of AI/ML and privacy and security. Dr. Luo has actively published in top conferences and journals such as IEEE S&P, ACM CCS, USENIX Security, NDSS, ACM Multimedia, IEEE TKDE, IEEE TIFS, IEEE TDSC, etc. He received the KU EECS Excellence in Undergraduate Teaching Professorship in 2023, the Miller Scholar award of University of Kansas in 2016, 2017, and 2021, and the Miller Professional Development Award in 2015. He is also the recipient of ACSAC 2017, ACSAC 2021, ACSAC 2025, and ACM/IEEE ICPC 2024 best paper awards, and CCS 2022 best paper honorable mention.
A/Prof. Neil Gong
Associate Professor, Duke University
Title: Prompt Injection in LLM Agents
Abstract: LLM is inherently vulnerable to prompt injection. In this talk, we will formally define prompt injection and illustrate it through example attacks, including stealing system prompts in LLM-integrated applications, inducing malicious tool selection in LLM agents, and manipulating web agents. We will then discuss defenses that detect whether an input has been contaminated by injected prompts. When contamination is detected, our approach further localizes the injected content, enabling applications such as post-attack forensic analysis and data recovery.
Bio: Neil Gong is an Associate Professor in the Pierre R. Lamond Department of Electrical and Computer Engineering at Duke University. His research interests lie in cybersecurity, with a recent focus on AI security and safety. He has received several honors, including the NSF CAREER Award, Army Research Office Young Investigator Program (YIP) Award, Rising Star Award from the Association of Chinese Scholars in Computing, IBM Faculty Award, Facebook Research Award, and Cisco Research Award. His work has also been recognized with multiple paper awards, including a Distinguished Paper Award at the IEEE Symposium on Security and Privacy and a Distinguished Paper Award Honorable Mention at the ISOC Network and Distributed System Security Symposium. He earned his Ph.D. in Computer Science from the University of California, Berkeley in 2015, and his B.E. from the University of Science and Technology of China in 2010 (with the highest honor).