Advanced Topics in Computer Science: AI Safety and Alignment

What safety risks do modern AI systems and methods pose? How can we mitigate these risks and ensure that AI systems are safe and aligned with the intention of their builder? What is currently being done to ensure that AI systems are safe? Are current safeguards sufficient? If not, how can we improve upon them? These are the questions we consider in this advanced seminar. Topics include algorithms and optimization-based methods for alignment (including reinforcement learning), sociotechnical AI safety, accountability and governance, among others.