Part 5: Secure-by-Design AI and MLSecOps

In this part, you will learn how to incorporate the attacks and mitigations we learned into a secure-by-design methodology, bringing security from the outset to AI development. You will learn about standard AI taxonomies from NIST, MITRE, and OWASP, threat modeling, and the use of security controls. You will understand how AI security relates to safety and ethics as part of Trustworthy AI. You will learn the principles and patterns of MLSecOps and how to apply these patterns with examples, using Jenkins, MLflow, and Python. Finally, you will cover how to mature and scale AI security beyond a single project with governance, as well as how to connect it with existing enterprise security.

This part has the following chapters:

• Chapter 17, Secure by Design and Trustworthy AI
• Chapter 18, AI Security with MLSecOps
• Chapter 19, Maturing AI Security