Nvidia CEO Jensen Huang concluded a pivotal two-day visit to Tokyo on July 16, solidifying a series of landmark agreements that position Japan at the forefront of the "physical AI" era and underscore the nation’s strategic drive for technological sovereignty. This high-profile engagement, following recent keynotes in Taiwan and a significant visit to South Korea, saw Huang courting Japan’s industrial titans and critical chip-supply chain elite, culminating in deals spanning a national AI factory, extensive partnerships with leading robotics companies, and deepened ties with material suppliers vital for Nvidia’s next generation of AI chips. Huang’s message was unequivocal: the next chapter of artificial intelligence will unfold on factory floors, powered by robots and intelligent machines, and he envisions Japan as the primary architect of this future. Many of the country’s largest manufacturers have swiftly aligned with this vision, recognizing a transformative opportunity.
This latest collaboration marks a full-circle moment in the intertwined histories of Nvidia and Japan’s industrial powerhouses. Three decades prior, a crucial $5 million investment from Sega proved instrumental in sustaining a then-struggling Nvidia. Today, the relationship has evolved into a mutual dependency, with both Nvidia and Japan’s industrial giants recognizing the imperative to co-create the physical AI era. This strategic alliance is set to commence with three cornerstone projects, designed to redefine manufacturing, logistics, and national technological capabilities.
Japan’s Sovereign AI Play: The Birth of Noetra and the Vera Rubin AI Factory
At the heart of Japan’s renewed industrial strategy is Noetra, a bold initiative to establish national AI infrastructure and models, reducing reliance on foreign-developed AI, particularly from the United States and China. Japan’s government has mobilized approximately 44 domestic firms, with industry heavyweights such as SoftBank, Sony, NEC, and Honda forming the core consortium. This collective endeavor is tasked with developing homegrown AI specifically tailored for robots, autonomous vehicles, and advanced factory floor operations. Tokyo has committed a substantial investment of up to 1 trillion yen (approximately $6.2 billion USD) over the next five years, signaling a profound national commitment to "physical AI" – foundation models engineered explicitly to command and operate machines in the real world. Japan’s ambition is clear: to own the software brain of its future industrial landscape.
While the software development remains domestic, the hardware foundation for this monumental undertaking will be provided by Nvidia. The U.S. chip giant is contracted to construct what it terms a "Vera Rubin AI factory," a colossal data center poised to become the world’s first national AI infrastructure. This state-of-the-art facility, slated for launch in 2028, will be packed with Nvidia’s cutting-edge next-generation chips, featuring an unprecedented deployment of 13,750 Vera CPUs and 27,500 Rubin GPUs, delivering an immense 140 megawatts of computational power. Noetra will oversee the development and operation of this critical data center, ensuring it serves as the central hub for training and deploying Japan’s sovereign AI models.
Noetra’s strategic roadmap is meticulously phased, outlining a progressive development pathway for its AI models. The first stage, expected in fiscal year 2026, focuses on a reasoning model heavily optimized for Japanese-language skills, crucial for domestic communication and nuanced understanding of local contexts. By 2028, the initiative plans to unveil an omni-modal version, capable of processing and understanding diverse data inputs including text, images, video, and audio, enabling more comprehensive environmental awareness for machines. The ultimate goal, targeted for 2030, is the release of "Real-world Native AI," models specifically designed and robustly tested to operate and control complex robotic systems in dynamic physical environments. These advanced models will be progressively released to external Noetra developers, fostering a broader ecosystem of innovation.
The Robotics Coalition: Japan’s Industrial Giants Embrace Nvidia Cosmos
Nvidia’s strategic focus on Japan’s factory floor has garnered enthusiastic support from the country’s most prominent robotics and manufacturing players. A formidable coalition of industrial giants, including Fanuc, Yaskawa, Kawasaki Heavy Industries, Fujitsu, Hitachi, NEC, Sony, SoftBank, Kubota, and the robotics industry group AIRoA, has publicly declared their intent to build upon Nvidia’s Cosmos models. Cosmos, an open-model initiative launched by Nvidia in May with several global AI labs, aims to create a unified framework for physical AI development.
In Tokyo, Nvidia provided a compelling incentive for these companies to commit: the unveiling of Cosmos 3 Edge. This specialized version of the Cosmos model is designed to run directly on Nvidia’s Jetson Thor chips, embedded within the machines themselves. This capability is crucial for enabling real-time, on-device AI processing, allowing robots to make instantaneous decisions and adapt to their environments without constant reliance on cloud connectivity. This advancement is particularly significant for Japan, a nation with a rich heritage in precision manufacturing and a global leadership position in industrial robotics.
Already, some partners are actively testing a shared control system leveraging Cosmos 3 Edge, streamlining operations across different robotic platforms. Others, such as Honda R&D and Omron, are rapidly integrating these new tools into their development pipelines. Jensen Huang articulated the profound implications of this shift: "The next frontier of AI is in the physical world, and this is a once-in-a-generation opportunity for Japan." He further emphasized, "Japan invented modern manufacturing. Now, it has the opportunity to reinvent it for the age of intelligent industries." This sentiment resonates deeply within Japan, where the imperative to innovate and maintain global competitiveness is paramount.
Toyota’s Expanded Partnership: Driving Physical AI from Road to Factory
The automotive sector, a cornerstone of Japan’s industrial might, is also deeply entwined with Nvidia’s physical AI vision. Toyota, a global leader in vehicle manufacturing, already utilizes Nvidia chips extensively across various layers of its technological stack. At CES in January 2025, Toyota committed its next-generation vehicles to Nvidia’s Drive platform, a comprehensive AI computing platform for autonomous driving and advanced driver assistance systems. This latest expansion of their partnership extends Nvidia’s reach into Toyota’s manufacturing processes, where advanced simulations are used to design and optimize production lines, enhance efficiency, and reduce costs.
Furthermore, Nvidia’s technology is being integrated into the sophisticated software that powers Toyota’s vehicles, enabling advanced features and capabilities. The collaboration also extends to systems designed to analyze and interpret road traffic data, contributing to safer and more efficient transportation. Toyota’s current strategy for advanced driver assistance systems (ADAS) adopts a more conservative approach compared to companies like Waymo and Tesla, which are pursuing fully autonomous systems with minimal human intervention. Toyota’s ADAS focuses on steering and braking assistance, still requiring an active human driver. This cautious yet technologically advanced strategy reflects Toyota’s emphasis on safety and gradual integration of autonomous capabilities. This extensive partnership with Toyota exemplifies the broad applicability of physical AI, moving beyond individual robots to entire fleets of intelligent machines on roads and in factories.
Why It Matters: Japan’s Strategic Imperative and Global Ambitions
Jensen Huang’s visit was more than a series of business deals; it was a pivotal moment that firmly embedded physical AI at the core of Japan’s national industrial strategy. Tokyo’s significant financial commitment underscores the urgency and strategic importance of this direction. Facing one of the world’s most acute demographic challenges, characterized by a rapidly shrinking and aging workforce, Japan views AI and robotics not as a luxury but as an existential necessity. Projections indicate that Japan aims to deploy 10 million AI-equipped robots across 18 critical sectors by 2040, a monumental shift in its labor landscape. This ambitious goal is backed by an estimated $65 billion in public and private investment dedicated to physical AI by 2040, demonstrating a whole-of-nation approach.
The longer-term vision is even more expansive. Japan’s AI Robotics Strategy, unveiled in March, sets an ambitious target to capture more than 30% of the global AI robotics market by 2040. Tokyo estimates this market to be worth approximately ¥20 trillion (about $133 billion USD), highlighting the immense economic opportunity at stake. The Ministry of Economy, Trade and Industry (METI) is actively funding the development of a domestic foundation model specifically designed to run these intelligent machines. The Noetra-managed, Nvidia-powered factory will serve as the crucial training ground for these models, enabling the development of truly massive AI systems with trillions of parameters, essential for complex real-world operations. Japan’s wager is that its unparalleled factory-floor data and its formidable manufacturing base will provide a unique competitive advantage in the race for physical AI, akin to how Silicon Valley’s internet data fueled the rise of digital AI.
Beneath the industrial and economic rationale lies a profound sovereign imperative. As the United States and China continue to accelerate their advancements in large-scale AI, Japan is acutely aware of the need to cultivate its own data, secure its own compute resources, and lessen its dependence on foreign-controlled infrastructure. This desire for strategic autonomy was evident during Huang’s visit, where he appeared alongside Trade Minister Ryosei Akazawa at the government’s official physical AI launch event on July 16, with Prime Minister Sanae Takaichi joining via video conference. The Takaichi administration has made AI and semiconductors central pillars of its broader economic growth plan, targeting ¥370 trillion ($2.3 trillion USD) in public and private investment across 17 strategic sectors by 2040. The Noetra factory, which Nvidia proudly bills as "the world’s first national AI infrastructure," stands as the clearest and most significant manifestation of this strategic bet. While Japan’s drive for independence is clear, for the foreseeable future, its foundational technological capabilities in advanced AI will undeniably rest on American silicon.
Working the Whole Room: Huang’s Strategic Engagement
Jensen Huang’s two-day itinerary in Japan was a masterclass in strategic engagement, reflecting Nvidia’s comprehensive approach to ecosystem building. He held private luncheons with virtually every key figure in Japanese technology, including the CEOs of Toyota, Fanuc, Yaskawa, Fujitsu, and Kawasaki Heavy Industries. Beyond the C-suite, Huang also dedicated time to connect with dozens of critical supply-chain chiefs over more informal settings, such as skewers and whisky in a Kanda izakaya. This dual approach of high-level executive meetings and grassroots engagement with the supply chain ensures that Nvidia fosters relationships across the entire industrial spectrum, from strategic visionaries to the practical implementers.
This "working the whole room" playbook is a hallmark of Huang’s recent global tours. Weeks prior, he delivered a highly anticipated homecoming keynote in Taiwan, engaging with the island’s crucial semiconductor ecosystem. Last fall, his visit to Seoul included a memorable fried chicken dinner and culminated in a massive 50,000-GPU deal, significantly expanding Nvidia’s ties with South Korean tech giants like Hyundai, Samsung, SK, and Naver. In each destination, Huang meticulously tailored his message and engagements to the specific industrial strengths and strategic priorities of the host nation. In Tokyo, the focus was laser-sharp on robots, the intricate supply chain, and the foundational chips powering this next industrial revolution. His visit underscored Nvidia’s commitment not just to selling chips, but to co-creating the future of intelligent industries alongside its global partners. The profound implications of these agreements are set to resonate across Japan’s economy and potentially reshape the global landscape of physical AI for decades to come.
