Threat Modeling: Designing for Security in an AI World, 2nd Edition
Threat Modeling: Designing for Security in an AI World will be available February 2, 2027. Help us get the word out!
About
Thirteen years after the original Threat Modeling: Designing for Security was released, there's a new version updated for a wholly different landscape. No matter who you are or what you're building, this book will show you how to bring threat modeling into your security development lifecycle and your broader design process. It's written for software developers, systems architects, security engineers, penetration testers, and IT administrators with security responsibilities alike. It's for professionals working on AI software, software developed with or by AI, or systems where AI never enters the picture.
Author and security expert Adam Shostack draws on decades of experience to lay out, in a way no other book does, how to build security into the design of software, services, and systems from the very beginning and how to do that in a world where AI is now part of the technology stack you're defending.
- Find and fix security issues before they hurt you or your customers
- Learn to use practical and actionable tools, techniques, and approaches for software developers, IT professionals, and security enthusiasts
- Identify and validate system boundaries, map attacker progression with proven kill-chain models, and choose the right defences for anything from a napkin sketch to a system that's been running since the first edition
- Explore software-centric threat modeling and see how it applies across the build phase and beyond, including in LLM-based and other AI-enabled systems
- Apply threat modeling to improve security when managing complex systems (or even simple ones!)
- Manage potential threats using a structured, methodical framework
- Discover and keep pace with evolving security threats, including those introduced by AI
- Draw on specific, actionable advice that holds up regardless of software type, operating system, or program approaches and techniques validated and proven to be effective at Microsoft and other top technology companies
Threat Modeling: Designing for Security in an AI World is full of tested, actionable advice for security professionals contributing to secure software development. From the first chapter, it teaches you how to use models to predict and prevent problems before you've written a line of code. The second edition carries the grounded approach of the first edition forward while addressing the design questions AI systems raise, so the guidance stays relevant as the technology you're building on changes.
Threat Modeling: Designing for Security in an AI World is as jargon-free and accessible as the 2014 classic. Written for everyone building and maintaining secure products, it hands you structured, repeatable methods for finding threats and designing defences across any technology stack. It ships on the same promise that made the first edition a classic: frameworks built for real projects on tight schedules. The first edition proved secure by design was possible. This one shows you how to make it happen.
Inside the new edition
- Review the Table of Contents. The page also includes chapters from the first edition that have been removed from the updated book.
Resources
Threat modeling: We have lots of threat modeling resources for your use.
Professors: Adam is working on an updated version of the instructor companion package so check back in the coming months, or use our Get in touch form. Wiley can provide copies for educators considering using the book as a textbook. Please use Wiley's textbook request form to request educator copies.