Jeffrey Ladish is an American AI security researcher and the founder and executive director of Palisade Research, a nonprofit that investigates dangerous capabilities of AI systems. Before founding Palisade Research, he worked on security at Anthropic.
His work focuses on demonstrating AI hacking capabilities, control-evading behavior, and loss-of-control risks through experiments and communicating the findings to policymakers. He speaks publicly on the pace of frontier AI development and safety measures.