SAN FRANCISCO – 30 Aug, 2026 : NVIDIA is rapidly expanding its ambitions beyond artificial intelligence in computers and data centres, positioning itself at the centre of a global race to develop humanoid robots capable of working alongside — and eventually replacing humans in — factories, warehouses and other industries.
The chip giant’s strategy is increasingly focused on what it calls “physical AI” — artificial intelligence that can understand and interact with the physical world.
NVIDIA is building an ecosystem that combines powerful processors, robotics software, simulation platforms and foundation models designed to give robots the ability to see, reason, learn and act.
One of its most important technologies is Isaac GR00T, a family of foundation models designed specifically for humanoid robots. The company is also promoting its Jetson Thor computing platform, which is designed to provide robots with the processing power required to run sophisticated AI models locally.
The strategy received a major boost through NVIDIA’s collaboration with South Korea’s LG Electronics to develop a next-generation bipedal humanoid robot.
LG plans to use NVIDIA’s Isaac GR00T models and Jetson Thor technology, while building a facility in Seoul intended to generate massive quantities of robot-training data. The companies are targeting a launch in the first quarter of 2027.
NVIDIA
For NVIDIA, the significance goes well beyond a single robot.
The company wants to become the technology platform underneath an entire generation of machines — supplying the computing hardware, AI models, simulation environments and software used by robot manufacturers around the world.
China is emerging as one of the most important arenas in that race.
More than 300 companies showcased robots at the World Robot Conference in Beijing this month, highlighting the extraordinary pace of development in China’s robotics industry. China now has more than 150 companies involved in humanoid robotics, according to Reuters.
Chinese manufacturers are increasingly experimenting with robots designed for industrial applications, logistics and other commercial uses. Many rely on NVIDIA technology during development and training, although US restrictions on advanced semiconductor exports are encouraging Chinese companies to develop domestic alternatives.
The technological progress has been striking.
At the World Humanoid Robot Games in Beijing, a Chinese humanoid robot reportedly completed a 100-metre race in 8.86 seconds — faster than Usain Bolt’s 9.58-second human world record.
But the robot subsequently crashed, underlining one of the industry’s biggest challenges: impressive demonstrations do not necessarily translate into reliable machines capable of operating independently in the real world.
Robots still struggle with balance, dexterity, perception and decision-making in unpredictable environments.
NVIDIA nevertheless believes the underlying AI technology is advancing rapidly enough to transform robotics.
AI boom
The company’s enormous data-centre business provides the financial firepower for that expansion. NVIDIA recently reported quarterly revenue of US$96.2 billion, more than double the previous year’s figure, with data-centre revenue accounting for the overwhelming majority.
The company is effectively using the AI boom to finance its next major technological bet.
The stakes are enormous. If NVIDIA succeeds, it may not need to manufacture the majority of the world’s robots. Instead, it could provide the “brains” and software infrastructure that power them.
That would transform NVIDIA from an AI-chip leader into one of the most influential technology companies in the emerging physical-AI economy.
The next great technology platform war may therefore not be fought over smartphones or cloud computing.
It could be fought over who controls the intelligence inside the world’s robots.
Leave a Reply
You must be logged in to post a comment.