Machine Learning (ML) has rapidly become a fundamental technology that underpins count- less applications, from natural language processing and computer vision to fraud detection and personalized recommendations. In recent years, there has been a growing understanding of how to use ML in security contexts, leading to the development of advanced tools and techniques for detecting and preventing malicious activities. However, the security and privacy aspects of ML itself remain less understood, posing new challenges and opportunities for researchers and practitioners. This Cybersecurity Body of Knowledge (CyBoK) Knowledge Guide (KG) aims to define the scope of adversarial machine learning and privacy in ML and provide an overview of the state- of-the-art in these rapidly evolving fields. Our focus is on the key challenges, open problems, and promising solutions that have emerged in the context of securing and preserving the privacy of ML systems. We speak with CyBOK Security and Privacy of AI authors Lorenzo Cavallaro and Emiliano De Cristofaro for an overview of the topic.