This episode discusses three critical topics in AI: the true nature of recent AI agent "breakouts," arguing they highlight governance and security flaws rather than model danger; the role of AGI narratives in fueling the current AI investment bubble and questioning its sustainability; and China's aggressive strategy in the global robotaxi market, potentially outpacing Western counterparts like Waymo.
Waymo co-CEO Dmitri Dolgov outlines seven crucial lessons from fifteen years of developing and scaling the Waymo Driver, the world's most advanced physical AI. He details the unique challenges of physical AI compared to digital, emphasizing the critical role of reliability, strategic technology choices, continuous innovation through foundation models, structure-augmented learning, high-fidelity simulation, AI flywheels, and robust evaluation frameworks to achieve superhuman safety and build trust in real-world autonomous systems.
Astro Teller, Captain of Moonshots at X, and Google co-founder Sergey Brin discuss the philosophy of radical innovation, X's unique culture of embracing failure and pressure testing ideas, and their reflections on past moonshots like Google Glass, Wing, and Waymo. They delve into the exponential nature of AI, its historical context, and the challenges of anticipating its future impact, while also highlighting crucial unsolved "white whale" moonshots for humanity, including education, battery technology, and redefining the intrinsic value of work beyond monetary compensation.
Dmitri Dolgov, co-CEO of Waymo, discusses the 20-year journey from the DARPA challenge to full autonomy. He explains the Waymo Foundation Model—a multimodal world action model powering the driver, simulator, and critic—and how their "end-to-end plus" architecture enables superhuman safety and exponential scaling.
A discussion on the evolving economics of autonomous vehicles, driven by end-to-end AI, and the growing local opposition to AI data centers due to concerns over resources like water, electricity, and noise.
Waymo Distinguished Engineer Vincent Vanhoucke discusses the core challenges of autonomous driving, explaining how Waymo fuses data from cameras, LiDAR, and radar to build a robust perception system. He delves into the "closed-loop" problem, the critical role of generative AI and simulation in training and validation, and how modern multimodal models are used in a teacher-student framework to distill vast world knowledge into the vehicle's onboard system, aiming for a safety standard that surpasses human performance.