Chip technology company Arm Holdings has announced a structural overhaul, creating a new unit dedicated to physical AI as interest in the sector continues to ramp up.
“The unit will look to develop technologies that combine AI and real-world physical movement, from vehicles to robots to autonomous machines,” an Arm spokesperson told AI Business.
The news comes as interest in physical and embodied AI continues to boom, with a spate of companies such as Nvidia making investments into developing the technologies.
Physical AI is defined as systems that can effectively operate in and respond to their physical environments, with particular use cases in the growing fields of humanoid robotics and autonomous vehicles.
The demand for these physical systems means companies are increasingly turning to predictable, low-latency systems that Arm said it is uniquely positioned to provide.
“Arm delivers both unmatched energy efficiency and the world’s largest software developer base, making it the natural platform for building and scaling physical and edge AI systems globally,” the company said in a blog post on the department’s launch. “That maturity means innovation can move faster, scale more broadly, and deploy more safely as AI transitions from digital intelligence to physical intelligence in the real world.”
Alongside physical AI, the company unveiled business units dedicated to cloud AI and edge AI.
“AI is transforming every industry and reshaping what’s possible,” the spokesperson said. “To stay aligned with these opportunities, Arm has evolved how we organize our business around where AI is happening: in the cloud, at the edge and in the physical world.”
The Cloud AI unit will focus on supporting high-performance AI across data center compute, while the Edge AI unit will make on-device AI across smartphones, wearables and other embedded devices its focus.
Arm-based chips are already used by several tech giants. The company’s compute platform is also used in Nvidia’s physical AI stack to accelerate AI that can “reason, plan, and adapt in dynamic environments,” according to the blog post.
The stack is being used to power robots and autonomous machines from companies including Boston Dynamics, LG Electronics and Neura Robotics.