Cybersecurity expert shares real-world lessons on AI security risks

Aonan Guan, a cybersecurity researcher and security engineer at Wyze Labs, recently served as a guest lecturer in AUM’s Department of Computer Science & Computer Information Systems.

During his lecture, titled “Security Risks in Modern AI Systems: Lessons from the Real World,” Guan walked attendees through case studies drawn from vulnerability disclosures, illustrating how security flaws emerge in modern agentic AI systems and the practical steps researchers and engineers take to responsibly disclose and remediate them. A lively Q&A session followed, drawing questions from both undergraduate and graduate students.

Guan has built a reputation for uncovering critical vulnerabilities in widely used platforms and AI systems, including Microsoft Azure, GitHub, Google Gemini CLI and Anthropic Claude Code. His research has been featured in The Hacker News, The Verge, Tom’s Guide, ITPro and other international media outlets. He is also an invited speaker at BSidesSF and Black Hat, two of the most prominent conferences in the cybersecurity industry.

The lecture was hosted by Pengzhou He, an assistant professor in the Department of Computer Science and Computer Information Systems. The department regularly hosts industry professionals and researchers to connect students with the evolving demands of careers in cybersecurity, computer science, software engineering and related fields.