My research sits at the intersection of control theory?The mathematical study of how to make complex systems behave safely and reliably. and artificial intelligence. With a background in aerospace engineering, I have seen firsthand the power of control systems to ensure safety and enhance performance across complex engineering systems. For example, once unimaginable, flight has become safe and routine today. My vision is that, similarly, control theory holds great power to ensure safety and enhance performance in the new class of intelligent systems we are witnessing with AI.

In the work I direct at the Control and Intelligent Systems Lab at UC Berkeley, we design Controllable AI: intelligent systems that we can principled design and reliably control, so they do what we want them to do, and never do what we don't want them to do. We pursue this by adapting mathematical principles from control theory to understand, control, and ultimately improve the behavior of AI systems, ranging from language models to robotic systems. If you want to learn more about this work, visit our website and connect with us here!