AI ALIGNMENT

Course Description

Artificial intelligence systems are rapidly becoming more capable and integrated into our society, raising urgent questions about how to ensure they behave in ways that are safe and aligned with human intentions. This course introduces the emerging field of AI alignment, the study of how to design safer AI models that are robust, interpretable, controllable, and ethical. Students will explore common failure modes in today’s systems, examine current approaches to developing safer AI models, and consider the broader social and ethical implications of AI progress and alignment research. The course will emphasize conceptual understanding and critical thinking rather than technical AI mastery, making it accessible to students from a variety of backgrounds. Prior coursework in AI/ML is helpful but not required.

Required Text

All assigned readings will come from research papers and articles provided by the instructor.