AI security focuses on protecting machine learning systems from manipulation and abuse, including data poisoning, evasion attacks, model theft, and adversarial inputs that can cause unsafe or unreliable behavior. You will find two complementary resources here: a free, audio-first course that teaches AI security concepts and how attackers actually target ML pipelines and deployed models, and a companion book on Adversarial Machine Learning that goes deeper with structured threat patterns and practical defensive techniques. The audio course is not an audio version of the book, it’s a standalone learning experience for building intuition and judgment, while the book is the focused reference you can use when you’re ready to evaluate risk and harden real systems.
AI Security
AI Security Course
The AI Security & Threats Audio Course is a comprehensive, audio-first learning series focused on the risks, defenses, and governance models that define secure artificial intelligence operations today. Designed for cybersecurity professionals, AI practitioners, and certification candidates, this course translates complex technical and policy concepts into clear, practical lessons. Each episode explores a critical aspect of AI security—from prompt injection and model theft to data poisoning, adversarial attacks, and secure machine learning operations (MLOps). You’ll gain a structured understanding of how vulnerabilities emerge, how threat actors exploit them, and how robust controls can mitigate these evolving risks.
The course also covers the frameworks and best practices shaping AI governance, assurance, and resilience. Learners will explore global standards and regulatory guidance, including NIST AI Risk Management Framework, ISO/IEC 23894, and emerging organizational policies around transparency, accountability, and continuous monitoring. Through practical examples and scenario-driven insights, you’ll learn how to assess model risk, integrate secure development pipelines, and implement monitoring strategies that ensure trust and compliance across the AI lifecycle.
Listen to the Trailer
Adversarial Machine Learning is a definitive guide to one of the most urgent challenges in artificial intelligence today: how to secure machine learning systems against adversarial threats. As AI moves from research into production, models increasingly influence decisions, automate workflows, and operate in hostile environments where attackers can probe, manipulate, and exploit them. This book frames adversarial machine learning (AML) as a practical security discipline, focused on protecting outcomes, maintaining trust, and ensuring that ML-enabled systems behave reliably when the inputs and operating conditions are not friendly.
The book explores the full lifecycle of AML, providing a structured, real-world understanding of how models can be compromised and what can be done about it. It walks readers through each phase of the machine learning pipeline, showing how weaknesses emerge during data collection and labeling, training and tuning, deployment and integration, and live inference. It breaks adversarial threats into clear categories based on attacker goals, whether to degrade availability, influence or tamper with outputs, steal models, or extract sensitive information from data and predictions. With clarity and technical rigor, it dissects the tools, knowledge, and access attackers need, and it explains how small changes in assumptions, interfaces, and observability can turn a “safe” model into an exploitable one.
In addition to diagnosing threats, the book provides a robust overview of defense strategies, from adversarial training and certified defenses to monitoring, privacy-preserving machine learning, and risk-aware system design that treats the model as one component in a larger secure system. Each defensive approach is discussed alongside its limitations and trade-offs, including cost, performance impacts, operational complexity, and where defenses fail under adaptive adversaries. The result is a grounded playbook for engineers, security leaders, and practitioners who need to evaluate real AI risk, choose protections that match the threat model, and build ML systems that remain dependable under pressure.
Adversarial Machine Learning
Recommended Podcasts
Get in Touch!
Nothing we do is perfect, so your help is always appreciated!
