Nvidia's CEO, Jensen Huang, recently concluded a high-stakes visit to Tokyo, solidifying numerous agreements across Japan's advanced technology sectors. This strategic tour, following similar engagements in Taiwan and South Korea, highlighted Nvidia's ambition to integrate its AI technologies deeply into the industrial fabric of Japan. The outcomes include a monumental national AI factory project, comprehensive partnerships with Japan's premier robotics enterprises, and crucial alliances with suppliers of chip materials essential for Nvidia's next-generation AI processors. Huang's clear message was that Nvidia is actively engaging with Japan's manufacturing base, inviting the nation's leading industrial players to partake in the evolution of AI-driven factory automation and robotics. He emphasized that the future of AI lies in physical applications, with Japan poised to lead this transformation.
Three decades ago, a $5 million investment from Sega was instrumental in rescuing a then-struggling Nvidia. Today, a new symbiotic relationship is emerging between Nvidia and Japan's industrial powerhouses, aimed at pioneering the era of physical AI. This collaboration is centered around three key initiatives.
Japan is embarking on 'Noetra,' a sovereign AI project, driven by a desire to develop its own AI capabilities for factories and robots, reducing reliance on external technologies. This ambitious undertaking involves approximately 44 domestic companies, with prominent roles played by SoftBank, Sony, NEC, and Honda. The Japanese government has committed a substantial 1 trillion yen ($6.2 billion) over five years to foster indigenous 'physical AI'—foundation models specifically designed for operating machinery. Japan's objective is to possess proprietary AI software, while Nvidia will supply the necessary hardware. Nvidia is constructing a 'Vera Rubin AI factory,' a colossal data center slated for a 2028 launch, equipped with 13,750 Vera CPUs and 27,500 Rubin GPUs, delivering 140 megawatts of power. Noetra will oversee the development of this data center and has outlined a three-phase plan: a reasoning model focused on Japanese language proficiency by fiscal year 2026, an omni-modal version processing text, images, video, and audio by 2028, and a 'Real-world Native AI' for robotics by 2030, with phased releases to external developers.
Furthermore, a powerful robotics coalition is forming, with Japan's industrial giants aligning with Nvidia's Cosmos initiative. Recognizing Japan's strong manufacturing heritage, Nvidia is intensely focused on the nation's factory floors. Many of Japan's leading robotics and manufacturing firms, including Fanuc, Yaskawa, Kawasaki Heavy, Fujitsu, Hitachi, NEC, Sony, SoftBank, Kubota, and the robotics group AIRoA, have pledged to integrate Nvidia's Cosmos models. This open-model endeavor, initiated by Nvidia in May with various global AI labs, received a significant boost in Tokyo with the introduction of Cosmos 3 Edge. This specialized version of the model is designed to operate on Nvidia's Jetson Thor chips directly within machines. Several companies are already experimenting with a unified control system, while others, like Honda R&D and Omron, are actively implementing these tools. Huang underscored the significance of this development, stating that the physical world represents the next frontier for AI and a unique opportunity for Japan to redefine modern manufacturing for the intelligent industry age.
Toyota, a major automotive player, is also deepening its collaboration with Nvidia. The company extensively utilizes Nvidia chips across its technological infrastructure. Having committed its next-generation vehicles to Nvidia's Drive platform at CES in January 2025, Toyota is now expanding Nvidia's reach into its manufacturing processes, employing simulations for production line design, integrating Nvidia's software into vehicle operations, and utilizing its systems for analyzing road traffic. Toyota's strategy emphasizes advanced driver assistance systems that manage steering and braking, maintaining a more cautious approach compared to fully autonomous solutions pursued by companies like Waymo and Tesla.
Huang's visit has undeniably elevated physical AI to a central position within Japan's industrial strategy, with the Japanese government backing this vision with substantial financial commitments. Confronted with a declining workforce, Japan aims to deploy 10 million AI-equipped robots across 18 sectors by 2040, supported by $65 billion in public and private investment in physical AI. The broader ambition is to capture over 30% of the global AI robotics market by 2040, a market Tokyo estimates to be worth approximately 20 trillion yen ($133 billion). The Ministry of Economy, Trade and Industry (METI) is funding a domestic foundation model for operating these machines, and Noetra's Nvidia-powered factory will be the hub for training large-scale models, potentially involving trillions of parameters. The strategic gamble is that Japan's rich manufacturing data and industrial base can replicate its past successes in the new domain of physical AI. Fundamentally, this industrial push is underpinned by a sovereign imperative. As the United States and China lead in large-scale AI, Tokyo seeks to secure its own data, computing power, and reduce dependence on external infrastructure. Huang's appearance alongside Trade Minister Ryosei Akazawa at the government's physical AI launch, with a video message from Prime Minister Sanae Takaichi, underscored this national commitment. The Takaichi administration has made AI and semiconductors cornerstones of its growth plan, targeting 370 trillion yen ($2.3 trillion) in public and private investment by 2040. Noetra's factory, hailed by Nvidia as the 'world's first national AI infrastructure,' represents the most tangible manifestation of this strategy, even if Japan's path to independence, for now, remains reliant on American chips.
