Anca Dragan

Anca Dragan is an associate professor in the University of California, Berkeley, Department of Electrical Engineering and Computer Sciences, currently on leave to lead AI Safety and Alignment at Google DeepMind. She focuses on enabling AI systems—such as robots, large language models and autonomous vehicles—to align with human goals and collaborate effectively with people. At UC Berkeley, she directs the Interactive Agents and Collaborative Technologies Laboratory (InterACT), where her research addresses AI alignment by developing methods for learning human preferences from diverse feedback and ensuring safe and coordinated human-AI interaction. At DeepMind, Dragan oversees safety efforts for the Gemini models, focusing on alignment with human values, robustness and informed oversight. She co-founded the Berkeley AI Research (BAIR) Lab, is a co-PI at the Center for Human-Compatible AI, and has consulted for Waymo on learning-based safety-critical systems. Her work bridges foundational AI research with practical safety and alignment applications.