Physical AI is having its moment–and everyone wants a piece of it

AI News

by Dashveenjit Kaur

March 4, 2026

AI-Generated Deep Dive Summary

Physical AI is experiencing a significant surge in momentum as multiple technologies converge, marking a pivotal moment for its adoption. Unlike traditional AI, which primarily processes data or generates content, Physical AI involves systems that perceive, reason, and act in the real world—such as robots, autonomous vehicles, and adaptive machines. This shift has been likened to the impact of ChatGPT on robotics, signaling a move from research environments to mainstream commercial use. In the West, major infrastructure companies are driving the development of Physical AI platforms. Nvidia, for instance, has introduced new models like Cosmos and GR00T for robot learning and reasoning, alongside energy-efficient hardware like the Jetson T4000 module. Similarly, Arm has established a dedicated business unit for semiconductor design in robotics and intelligent vehicles. Siemens and Nvidia are collaborating to build an Industrial AI Operating System aimed at creating fully AI-driven manufacturing sites. Google, too, is integrating its AI models, software, and cloud infrastructure to create a vertically integrated stack for manufacturers. In contrast, China is leading the charge in hardware development, accounting for over 80% of global humanoid robot installations in 2025. Companies like Alibaba are investing in open-source AI models (e.g., RynnBrain) to enhance robotic comprehension of the physical world. This progress is underpinned by China’s control of key components such as lidar sensors and harmonic reducers, as well as economies of scale that have reduced hardware costs. This convergence of technology and infrastructure has significant implications for industries worldwide. A Deloitte survey reveals that 58% of global business leaders are already using Physical AI, with plans to adopt it at an even faster pace over the next two years. The focus is now on how quickly—and

Verticals

Originally published on AI News on 3/4/2026