Description

This seminar explores critical topics in the rapidly evolving field of AI safety, focusing on challenges in and solutions to ensuring the reliable, ethical, and beneficial deployment of artificial intelligence systems. Each week we will have a leading expert from industry or academia discuss one aspect of AI safety. We will have speakers from leading AI companies such as OpenAI, Google DeepMind, AWS, and more to provide unique insight into how these organizations are practically addressing safety challenges. Topics range from developing safe and robust AI systems, model verification and validation, system monitoring and oversight, explainability of black-box models, ethical considerations, and emerging problems. Focus areas include natural language processing, autonomous vehicles, and societal impacts. The objective is for participants to receive broad exposure to different aspects of AI safety to foster a more complete understanding of the discipline.

Attendance

If you are taking the class for credit, you need to attend 7 out of 9 talks. Please use this form to record your attendance and let the course staff know if you have a conflict, illness, etc and cannot attend in person. No recording will be available since the course room is not outfitted to record.