Alan Chan - AI Alignment and Governance at NeurIPS

This article summarizes Alan Chan's research interests and background, focusing on AI alignment and governance. It highlights his work on measuring harms from language models, understanding agent incentives, and controlling values in machine learning models. The article also mentions his involvement in NeurIPS and the audio quality limitations of the discussion. The content is informative and provides a good overview of Chan's research.

• •Alan Chan is a PhD student at Mila, researching AI alignment and governance.
• •His research focuses on measuring harms from language models and controlling values in ML models.
• •He has worked on various projects related to AI ethics and governance, including explainability, scoring rules, and global exclusion.
• •The article is based on a discussion at NeurIPS.