Applied Intuition, a $15B physical AI company, discusses deployment of autonomous systems across vehicles, mining, construction, agriculture, and defense

Applied Intuition has built a physical AI platform spanning simulation and RL infrastructure, operating systems for vehicles and machines, and fundamental AI models for autonomy. The company powers cars, trucks, construction and mining equipment, agriculture, defense, and driverless L4 trucks running in Japan today.

Physical AI differs from screen-based AI because safety-critical machines like driverless trucks and autonomous vehicles require much higher reliability—learned systems can make mistakes in chat or coding, but failures in real hardware are far more costly.

The core bottleneck in autonomy is no longer model intelligence but deployment onto constrained hardware: onboard vehicle models need millisecond-level latency, low power, and small size, while data-center models can be huge and slow. Validation is shifting from binary pass/fail requirements toward statistical safety metrics measuring "how many nines" of reliability and mean time between failures.