Course Overview

Module 1. Foundations of Artificial Intelligence and Machine Learning Systems

1.1 What is artificial intelligence and machine learning

1.2 Categories of machine learning systems

• Supervised learning

• Unsupervised learning

• Reinforcement learning

1.3 Machine learning lifecycle: training versus inference

1.4 Model generalization, overfitting, and performance trade-offs

1.5 Data fundamentals: datasets, features, labels, and embeddings

1.6 Classical machine learning models versus large language models (LLMs)

1.7 Architectural trust boundaries in AI systems

1.8 Common AI/ML deployment patterns in production environments

Module 2. Introduction to AI and Machine Learning Security

2.1 Core components of AI/ML systems and application architecture

2.2 End-to-end ML pipeline overview

• Data acquisition

• Model training

• Model deployment

• Inference and consumption

2.3 Common AI/ML frameworks, APIs, and platforms

2.4 AI/ML-specific attack surfaces

2.5 Model-level versus application-level security risks

2.6 Overview of the MITRE ATLAS framework

2.7 Overview of the OWASP Top 10 for Large Language Model Applications

Module 3. Reconnaissance and Threat Modeling for AI Systems

3.1 Identifying AI and ML components in target applications

3.2 Discovering model usage through application behavior analysis

3.3 Model type and architecture fingerprinting

3.4 API endpoint enumeration and inference analysis

3.5 Identifying training data sources and update mechanisms

3.6 Threat modeling using MITRE ATLAS tactics and techniques

3.7 Mapping attack surfaces to attacker objectives

3.8 AI/ML security assessment methodology