Dr. Xudong Han is dedicated to advancing AI safety, reliability, and responsible AI practices through his work at LibrAI and as a Research Scientist at MBZUAI. His research focuses on post-training alignment, scalable oversight, and evaluation frameworks for large language models, with contributions to safety testing programs at OpenAI, Anthropic, and MBZUAI IFM.