Nvidia turns its AI momentum toward “physical AI,” pitching a nearly $500 billion robotics market

Nvidia’s AI growth story is entering a new phase, with the company increasingly tying its next leg of expansion to “physical AI,” the use of AI systems that act in the real world through robots, vehicles, factory equipment and other embodied devices. The push matters because Nvidia’s early lead in generative AI was largely built on GPUs and the infrastructure required to train and run models in data centers, where scale can be achieved quickly by adding computing capacity.

The company reported record results for its first quarter of fiscal 2027, ended April 26, 2026. Revenue was $81.6 billion, up 85% year over year, while Data Center revenue reached $75.2 billion, up 92% year over year. Nvidia also announced a large return of capital, including an additional $80.0 billion share repurchase authorization approved on May 18, and a dividend increase to $0.25 per share. In the same release, Nvidia said it is transitioning to a two-market reporting framework, separating Data Center and Edge Computing, with Edge Computing spanning devices for agentic and physical AI including PCs, robotics and automotive systems.

Nvidia’s accounting change is more than a re-labeling. It reflects a broader effort to reposition itself as a computing platform for AI across environments, not only a supplier to cloud data centers. In the quarter’s commentary, CEO Jensen Huang said agentic AI is scaling across companies and industries, and the company positioned itself as the platform that runs across hyperscale data centers to the edge. For investors, that shifts the question from whether Nvidia can keep selling chips to how much value Nvidia can capture as AI expands into robotics, industrial automation and other physical deployments.

The physical AI theme surfaced prominently at COMPUTEX 2026 in Taipei, where the show highlighted robotics and intelligent mobility and launched its inaugural AI Robotics Zone. A COMPUTEX-linked release citing Strategy& (part of PwC) said physical AI is expected to generate approximately €430 billion in global market value by 2030, with large-scale commercial adoption within the next three to five years across manufacturing, logistics, healthcare and aerospace. The same release said COMPUTEX 2026 drew 111,312 buyers and visitors from 152 countries and regions, underscoring that robotics and real-world AI are becoming core talking points for technology buyers, not side tracks.

Nvidia is backing the physical AI narrative with specific tooling for humanoid robots. On June 1, 2026, Nvidia announced the NVIDIA Isaac GR00T Reference Humanoid Robot for academic research, describing it as the first open humanoid reference design built on its Isaac GR00T platform. The reference design integrates a Unitree H2 Plus humanoid robot chassis, Sharpa Wave five-finger tactile hands for dexterous manipulation, and NVIDIA Jetson Thor onboard compute for advanced reasoning and control, combined with Isaac GR00T open software and models. Nvidia said the reference design is meant to reduce fragmentation across hardware integration, data collection, simulation, training, evaluation and deployment, and it named several universities and research labs expected to use the system.

The strategic logic for Nvidia is straightforward: physical AI requires more than raw GPU horsepower. It also depends on simulation, data capture workflows, software stacks for moving from training to deployment, and real-time on-robot inference. If robots and other embodied AI systems begin adopting standardized development platforms, Nvidia may be able to expand its footprint from selling accelerators to selling a larger toolkit that developers use to build, test and operate machines.

Still, the company has not put a number on how quickly it expects to monetize physical AI, and “nearly $500 billion” remains an industry market estimate rather than Nvidia guidance. Deploying AI in physical environments can be slower than ramping compute in data centers, and it comes with higher requirements around safety, reliability and validation in real-world settings. Nvidia’s challenge will be whether its robotics and edge computing strategy can translate into durable revenue streams alongside the still-dominant data-center cycle.

Looking ahead, investors will likely watch how Nvidia’s Edge Computing line trends in future quarters, along with indicates from robotics ecosystems around adoption of its Isaac GR00T platform and related simulation and deployment tools. It will also be important to see whether Nvidia’s physical AI messaging is matched by measurable customer commitments and real-world deployments rather than mostly demonstration-stage progress. The next test is timing, not only technology. Ultimately, the market opportunity can be large, but the pace of commercialization will determine what portion of it becomes revenue.