AI Security - From Models to Agents
AI stopped being something you talk to. Now it acts.
The moment a model can send an email, call an API, run code or move money, every weakness becomes a path to real consequences. This is the hands-on, lab-driven guide to defending it.
Why this book
From a single classifier to autonomous agents.
You will not just read about attacks — you build them against a target system that runs throughout the book, then build the defences. One coherent guide, written for the people who now have to secure AI.
What you get
Inside the book
The AI attack surface
Threat modelling with MITRE ATLAS, and the machine-learning supply chain that feeds every model.
Attacks on models
Adversarial examples and data poisoning — built, understood, then defended in full.
Securing agents
The security of autonomous, tool-using agents that send email, call APIs, run code and move money.
Get it now
Secure AI from models to agents.
Instant PDF, written by someone who has actually done the work. €44.99.
I want this — €44.99