AI for Cybersecurity: Research and Practice

AI for Cybersecurity: Research and Practice brings together current thinking on one of the most active intersections in modern computing: how AI can support security work, and where it introduces new risk. Edited by Houbing Herbert Song, Elisa Bertino, Alvaro Velasquez, Huihui Helen Wang, Yan Shoshitaishvili, and Sumit Kumar Jha, this first edition is framed as a practical reference for readers who want both the technical foundations and the larger security context.

What this book focuses on

The book is organized around four defensive capabilities — deter, protect, detect, and respond — and looks at AI as part of a broader cybersecurity strategy rather than a standalone tool. That makes the discussion useful for readers who want to understand how methods behave in real security settings, not just in theory.

Key topics covered

• robustness and adversarial machine-learning threats
• privacy risks such as model inversion and membership inference
• forensic and formal methods for auditing AI components
• generative AI systems in security use cases
• transparency, explainability, fairness, and bias

Designed for researchers and practitioners

This is the sort of book that speaks to advanced undergraduates, postgraduates, and working professionals who need a structured overview of the field. It is especially relevant for readers looking at cybersecurity from the perspective of AI design, deployment, evaluation, and governance.

A current view of an evolving field

Because AI and cybersecurity continue to move quickly, a title like this is valuable as a snapshot of the state of practice in 2026. It does not treat AI as magic; instead, it examines where AI helps, where it struggles, and what security teams and researchers need to consider when applying it responsibly.

For readers building knowledge in AI-driven security, this edition offers a focused, expert-led survey of the field with clear attention to both opportunity and risk.