- ホーム
- > 洋書
- > 英文書
- > Computer / General
Full Description
Cascading chaos, your agents are talking to each other creating unintended trust. Stop chained identity attacks before they start, by mastering the art of machine-to-machine interactions, to lock down your automated future.
Key Features
Understand how AI agents reshape identity risks across modern infrastructure
Implement zero-trust machine identity controls using open-source technologies
Modernize existing security practices to govern AI at enterprise scale
Purchase of the print or Kindle book includes a free PDF eBook
Book DescriptionPrevent AI agents from becoming your next security blind spot. As autonomous systems, integrations, and machine-to-machine communications multiply, traditional identity controls struggle to keep pace. This book helps you identify, govern, and secure non-human identities before they create opportunities for lateral movement, privilege abuse, and trust-based attacks across your infrastructure.
Written by cybersecurity leader Sean Juroviesky, this book provides a structured approach to securing AI-enabled environments. You'll understand how AI systems interact with infrastructure, why non-human identities introduce unique risks, and how to assess those risks through an architectural security lens.
You'll explore technologies such as SPIFFE, SPIRE, Open Policy Agent, service meshes, certificate authorities, and zero-trust architectures. You'll also strengthen existing controls through network segmentation, RBAC, ABAC, data masking, and tightly managed permissions. Rather than relying on isolated fixes, the book shows you how to build a sustainable machine identity strategy that scales with your AI initiatives.
By the end, you'll be able to design and operate secure AI ecosystems, implement effective guardrails for AI agents, and confidently manage non-human identities across cloud and enterprise infrastructure.What you will learn
Practical AI infrastructure use cases you can deploy today
What non-determination is and why it's needed in AI
What non-human identity is, and how it differs from human identity
The threats that AI agents present to your infrastructure
How to use a framework to discover AI risks and mitigate them
Strong guardrails you can implement for your AI uses
How to deploy modern identity tools, within your infrastructure
Familiar tools and how they can be adapted for securing AI
Who this book is forThis guide is essential for infrastructure engineers, security architects, cloud developers, and DevSecOps professionals navigating the AI boom. Designed for mid-level to advanced practitioners, this book takes you on a journey from managing legacy credentials to mastering autonomously controlled non-human identities. You will evolve your skillset, achieving the ultimate goal of: architecting resilient, AI workflows. By the end, you will confidently implement dynamic guardrails that secure your infrastructure at scale, without needing to slow down innovation.
Contents
Table of Contents
Wouldn't it be neat, promises of AI
Non-deterministic identities, and the threats it presents
Inherent vulnerabilities of AI
Security Escape, how young technologies escape controls
New frameworks, guardrails, and familiar tools, a strong step forward
The Presidio AI Framework, and NeMo Guardrails
Securing identity with SPIFFE, OIDC, and lifecycle management
Familiar methods create strong controls
The power of combined strategies, and adaptability of technology
The abstraction, AI is another layer not a unique threa
AI blackboxes create distrust, verifying what you can't see
Building and enforcing multiple layers of controls
Retaining hope despite the challenges AI presents



