Evaluator · round one
Independent AI researcher; formerly Google DeepMind and OpenAI Governance; Manifund regrantor.
$3.00M of Richard's recommendations funded
$2.56M granted through Richard's order
Richard Ngo is an independent AI researcher and philosopher who writes about the future of AI and society. He was previously a research engineer on the AGI safety team at Google DeepMind and then a member of OpenAI's Governance team, where he worked on forecasting the capabilities and risks of advanced AI. He studied computer science and philosophy at Oxford and machine learning at Cambridge. He was lead author of The Alignment Problem from a Deep Learning Perspective (ICLR 2024), and his sequence AGI Safety from First Principles is one of the most widely read introductions to the field.
Richard is an AI-safety regrantor on Manifund, where funders delegate budgets to experts who choose projects and publish their reasoning for each grant. He also designed the Alignment Fundamentals curriculum that BlueDot Impact has used to train many people now working in AI safety. His other writing, including the science-fiction collection The Gentle Romance, is on his website.
