Security of AI

Catalog description

This course introduces the challenges of securing modern AI systems, focusing on threats such as evasion, poisoning, backdoors, model theft, privacy attacks, and jailbreaks. Students will learn how these threats impact AI models in practice and study defenses to make AI more robust and trustworthy. The course combines conceptual and algorithmic foundations with hands-on exercises, enabling students to observe attacks and defenses directly across multiple types of AI systems, including traditional machine learning, deep learning, and large language and vision models.