Japan launched the FRONTia Project and partnered with Noetra Corp. and Nvidia to build a national physical AI infrastructure, deploying 13,750 Vera CPUs and 27,500 Rubin GPUs in a 140-megawatt data center.
Building the World’s First National AI Infrastructure for Physical Systems
Japan is expanding its national strategy for physical artificial intelligence through a government-backed computing infrastructure project and new collaborations with domestic manufacturers and robotics companies aimed at accelerating industrial automation and AI development. The initiatives include plans to establish what organizers described as the world’s first national AI infrastructure dedicated to physical AI and a broader effort by Japanese industry to develop intelligent machines using AI platforms.
Japan’s Ministry of Economy, Trade and Industry (METI) is supporting the FRONTia Project, a national program to develop multimodal foundation models for robotics and physical AI. The project will use an AI factory to be built by Noetra Corp. as its computing platform for developing AI agents, robotics, digital twins and other industrial AI applications. NVIDIA today announced it is working with Noetra Corp. to launch an NVIDIA Vera Rubin AI factory with 13,750 NVIDIA Vera CPUs and 27,500 NVIDIA Rubin GPUs for national physical AI.
The new AI factory, established by Noetra, will be architected with NVIDIA Vera Rubin NVL72 racks using the NVIDIA DSX™ platform, connected and scaled with NVIDIA Spectrum-X™ Ethernet networking. It will deliver 140 megawatts of data center capacity based on the NVIDIA DSX platform. The NVIDIA Vera Rubin AI factory will provide the computing foundation for Japan’s FRONTia Project, which refers to the project titled, Development of Multimodal Foundation Models with a View to AI Robotics and Physical AI,
launched by METI. The project brings together the country’s manufacturing expertise, real-world industrial data and global technology leaders to develop highly reliable multimodal foundation models for physical AI. The pretrained weights of Noetra’s multimodal foundation models will be made broadly available to domestic model developers and enterprises alongside software.
Japan has launched the FRONTia Project, which will serve as the core of the country's physical AI ecosystem,
METI Minister Ryosei Akazawa said. “By fostering collaboration between Japan and leading global innovators — including Nvidia — and leveraging Japan’s strengths, such as its onsite expertise and manufacturing technology infrastructure, we will build highly reliable multimodal foundation models and contribute to solving global social challenges,” he said.
Converting Frontline Industrial Expertise Into AI-Ready Data
The replacement of intellectual work in cyberspace by AI is advancing at an unprecedented pace, driven by the rapid evolution of generative AI. However, the primary arena of value creation is already shifting into its next phase: the domain of “physical AI”—systems that integrate multimodal data such as images, audio, tactile information, and various sensor inputs to understand the real world and execute physical tasks. The global race in AI development, particularly around large language models, has been driven by massive web datasets, enormous computing resources, and global cloud infrastructure. In this arena, global technology giants have established a clear lead.
By contrast, the social implementation of physical AI will not be determined by AI model performance alone. What matters is the ability to integrate hardware, sensors, control systems, safety protocols, and operational design—all while continuously improving systems in real-world environments. The axis of competition is shifting from the scale of models to the power of integration and operation that creates value in the field.
This shift may present a historic opportunity for Japan, a country that has long cultivated precision manufacturing and high-quality services. Yet Japan’s strength lies not merely in the operational excellence that underpin these capabilities. The essential question is whether Japan can transform the tacit knowledge accumulated on frontline workplaces into data structures that AI can learn from, evaluate, and continuously improve, and ultimately elevate that knowledge into reusable industrial infrastructure. Japan’s advantage in physical AI lies in converting its tacit frontline knowledge into AI‑ready data, redesigning workplaces for AI‑robot integration, and building secure domestic data infrastructure to sustain industrial competitiveness.
Broad Industrial Integration
Japanese manufacturers and technology companies are expanding work on physical AI using Nvidia’s Cosmos and Isaac platforms for applications in manufacturing, logistics, healthcare, telecommunications and more.
