QCraft CEO: autonomous driving leads path to physical AI

QCraft Chairman and Chief Executive Dr James Yu outlined a three-stage development arc for autonomous driving at the Intelligent Vehicles & Production 2026 conference in Garching bei München on 18 March, arguing it represents the most commercially viable pathway to physical AI. The two-day event, organised by the Center Automotive Research (CAR) and Technische Universität München (TUM), drew senior leaders from Mercedes-Benz, Audi, Bosch, Siemens and Rheinmetall.

Dr Yu described the field as having progressed from modular machine intelligence and end-to-end learning to a third phase: superhuman intelligence, driven by vision-language-action (VLA) large models, world models and reinforcement learning. In this phase, he argued, AI no longer imitates human driving but begins to understand the physical world.

More than one million vehicles now operate with QCraft’s Navigate on Autopilot system, forming what Dr Yu described as a large-scale training ground for physical AI. To address the cost and complexity of real-world testing, the company has built a virtual driving school using world models to simulate safety-critical scenarios before deployment.

“In the digital world, AI has already approached the level of general intelligence, and may even be entering the era of superintelligence. But the next great breakthrough will come from the physical world. When AI begins to understand gravity, friction, and human intention, that is where the biggest impact will be felt,” Dr Yu said.

QCraft opened its European headquarters in Munich in September 2025. Dr Yu added that the company’s physical intelligence platform is intended to extend beyond passenger vehicles to robots and other machines that must perceive, reason and act in the physical world.

Source: Business Wire