NVIDIA's Robotics Revenue Surges 72% Year-on-Year, CEO Predicts Billions of Robots

NVIDIA's CEO has predicted that robotics, particularly humanoid robots, will be the next major growth engine for the AI chip giant, following artificial intelligence. The company's robotics business, which includes autonomous driving, saw quarterly sales of approximately 567 million dollars in May, accounting for about 1% of the company's total sales. However, the CEO expects this segment to grow rapidly. "We have many significant growth opportunities across the company, with AI and robotics technology being the two largest, with a total market size of tens of billions of dollars," the CEO said at the annual shareholder meeting.

Over a year ago, NVIDIA significantly adjusted its business disclosure methods, merging the autonomous driving and robotics business units into a single category. In May, the company's financial report showed that the revenue of this business unit for the quarter was 567 million dollars, accounting for approximately 1% of total sales, but it increased by 72% year-on-year. Based on the technology roadmap disclosed by NVIDIA at events such as CES and GTC this year, global robotics and autonomous driving development is expected to increasingly adopt NVIDIA's full-stack development platform.

Through platforms such as Omniverse, Cosmos, Isaac Sim, NVIDIA DGX Cloud, and the NVIDIA humanoid robot hardware computing platform, NVIDIA aims to build a robot AI large model software and key component hardware ecosystem with the ability to simulate and emulate the physical world in all directions. The company is determined to become the core "foundation" in the fields of AI humanoid robots and fully autonomous driving. NVIDIA is committed to integrating its new NVIDIA robot technology, which simulates and emulates the physical world in 3D, into the robot model development and training process of technology companies, as well as verifying robot behavior, vision, and basic models similar to GPT in a simulated environment. The reliability and actual capabilities of the complete integrated robot AI large model logic in the real world are also being tested. The "cyberpunk world" seen in science fiction films and games such as Blade Runner 2049 and Cyberpunk 2077 seems to be getting closer, and AI humanoid robots will soon accompany billions of people on Earth until they grow old.

Although the current scale of the robotics business is relatively small, the CEO pointed out at the shareholder meeting that related robotics applications require dedicated data center AI chips to train robot large models and other software platforms. Additionally, high-performance chips need to be installed in autonomous vehicles and robot bodies. The CEO mentioned the NVIDIA Drive autonomous driving platform based on high-performance automotive-grade autonomous driving chips, which is already in use by Mercedes-Benz, as well as the recently launched AI large model platform Cosmos for AI humanoid robots.

"We are moving towards an era where the world will soon have tens of billions of robots, hundreds of millions of fully autonomous vehicles, and tens of thousands of super robot factories driven by NVIDIA's software and hardware ecosystem technology," the CEO said.

Over the past three years, driven by the explosive growth in demand for data center AI GPUs, NVIDIA's revenue has surged. NVIDIA's revenue scale has soared from approximately 2.7 billion dollars in the 2024 fiscal year to 13.05 billion dollars in the 2025 fiscal year, and it is expected that the revenue for the fiscal year ending in January 2026 will approach 20 billion dollars. As of the end of the first quarter on April 27, NVIDIA's overall revenue increased by 69% year-on-year to 4.41 billion dollars, while analysts' average estimate was 4.33 billion dollars. This growth rate is enviable for most chip companies, although it is the lowest percentage increase for NVIDIA in two years.

With the strong contribution of the Blackwell architecture AI GPU, the revenue of the data center business unit is approximately 3.91 billion dollars. The revenue scale of this business unit alone has already exceeded the total revenue of all of NVIDIA's recent AI computing competitors, including Intel and AMD. Revenue related to gaming, which was once NVIDIA's main business, is approximately 380 million dollars, with analysts' average prediction being 285 million dollars. The revenue from the automotive and robotics business is approximately 567 million dollars.

The CEO emphasized that NVIDIA is continuously launching complementary technologies for high-performance AI chips, including software development platforms, cloud computing services that include cloud-based AI computing power, and super-high-performance network chips to connect AI accelerator clusters. "We have long since stopped seeing ourselves as just a chip company," the CEO emphasized, adding that it is now more appropriate to refer to NVIDIA as a large provider of "AI infrastructure" or "computing platforms."

The CEO stated that the company is collaborating with hybrid cloud platform companies to build "full-stack AI factory infrastructure" to help enterprises release data value "at an unprecedented speed and accuracy." "We are entering a new era of industrialization characterized by large-scale generative AI intelligence," the CEO said.