On AI:AM

Max Nadeau

Program Officer, Technical AI Safety, Coefficient Giving

Max Nadeau funds technical AI safety research at Coefficient Giving, the grantmaker formerly known as Open Philanthropy. His team backs people and organizations working to make increasingly capable AI systems more trustworthy and understandable — alignment researchers, independent organizations auditing AI companies, and builders using AI to improve society's security and decision-making. He joined in October 2023; before that he studied computer science at Harvard and ran fieldbuilding at Redwood Research, including the Machine Learning for Alignment Bootcamp. His own research includes work on misleading visual features that cause image classifiers to fail, on locating the components inside a language model that retrieve variable values in arithmetic problems, and on measurement tampering — the problem of an AI making the evidence look successful without actually accomplishing the task.