Securing AI Using Zero Trust Principles

Zero Trust has become the dominant framework for modern network security, built on the principle that no user, device, or system should be trusted by default regardless of where it sits relative to a network perimeter. Securing AI Using Zero Trust Principles applies that framework specifically to AI systems, addressing the unique challenges that arise when the entities you’re trying to protect — and the threats you’re trying to defend against — include AI models, training pipelines, inference systems, and the data flows connecting them. Cindy Green-Ortiz, Zig Zsiga, and Saskia Schröer bring a networking and infrastructure background to the problem, which shapes the book’s emphasis on how Zero Trust architecture needs to adapt to accommodate AI workloads’ distinct patterns of data access, model serving, and inter-system communication. Published as part of Cisco Press’s Networking Technology: Security series, the book carries the technical credibility and depth that series is known for, working through implementation specifics rather than staying at a high-level conceptual overview. At 704 pages, it’s one of the most substantial treatments of AI security from an infrastructure and networking perspective, suited to security architects and network engineers who need to understand how Zero Trust principles translate into concrete design decisions for AI-enabled environments. For organizations serious about securing AI infrastructure at the network level, this offers rare depth on a topic most AI security resources don’t address with this degree of technical specificity.

Author:
Cindy Green-Ortiz, Zig Zsiga, Saskia Schröer