Joe Carlsmith

Joe Carlsmith works at Anthropic on the constitution for Claude, and writes independently about various topics in philosophy and futurism. He was previously a senior advisor at Coefficient Giving (formerly Open Philanthropy), where he focused on existential risk from advanced artificial intelligence and wrote reports on power-seeking AI and on whether AIs might fake alignment during training. His essay series Otherness and control in the age of AGI explores how agents with different values should relate to one another, and the ethics of seeking and sharing power. He has a doctorate in philosophy from the University of Oxford.