Title: AI in Cybersecurity: Attribution, Inference, and the Dynamics of the Attack?Defense Arms Race
Abstract: The rapid progress of artificial intelligence has ushered in new methodologies for modeling, learning, and inference across domains. In security, this evolution offers both promise and peril. This keynote explores the dual role of AI in the cybersecurity landscape–empowering defenders with advanced capabilities such as code authorship attribution, LLM-driven vulnerability description enhancement, automated repair, and in-network intelligence for self-driving networks, while also exposing novel vulnerabilities, including privacy breaches in emerging modalities such as AR/VR systems. Drawing from empirical studies and system-level research, we illustrate how AI reshapes the balance in the attack-defense arms race. The talk concludes with a discussion of open research questions and methodological challenges, calling for a deeper synthesis of AI robustness, interpretability, and trust in the design of future-secure systems.
Biography: David Mohaisen is a Professor of Computer Science at the University of Central Florida, where he directs the Security and Analytics Lab (SEAL). His research spans systems and security, with contributions in malware analysis, adversarial machine learning, privacy inference, blockchain security, in-network intelligence, and trusted hardware. His work appears in top venues such as IEEE S&P, USENIX Security, CCS, NDSS, HPCA, CHI, VR, and ICCV. Prior to UCF, he held research and faculty positions in both academia and industry, including at Verisign Labs, the University at Buffalo (SUNY), and ETRI in South Korea. He has led major research efforts funded by NSF, NRF, AFRL, and various industry partners. He has served on editorial boards for IEEE TDSC, TMC, TPDS, and TCC, and regularly contributes to the organization of leading conferences, including IEEE S&P, CCS, NDSS, etc. He is a senior member of ACM and IEEE, and a Distinguished Speaker (ACM) and Distinguished Visitor (IEEE).