Nvidia CEO Jensen Huang spent two days – July 15 and 16 – in Tokyo, courting the Japanese industry and chip-supply elite, weeks after a keynote speech in Taiwan, and months after visiting South Korea. He walked away with deals covering the entire Japanese tech ecosystem: a national AI factory, partnerships with the country’s leading robotics companies, and deals with chip-material suppliers that power the next generation of Nvidia chips. His message was clear. Nvidia is targeting a factory floor in Japan, and many of the country’s major manufacturers are joining. The next chapter of AI, Huang said, is on factory floors, robots, and machines, and he wants Japan to build it.
Thirty years ago, Sega’s $5 million investment helped keep Nvidia afloat; today, Nvidia and the Japanese industrial giants need each other again – this time to build the virtual AI era, starting with these three projects:
Noetra – Japanese AI game. The country does not want to use factories and robots for American or Chinese AI. Therefore, the government has brought together around 44 domestic firms, with SoftBank, Sony, NEC and Honda at the core, to develop its own AI for robots, cars and factory floors. Tokyo is making up to 1 trillion yen ($6.2 billion) in revenue over five years, betting on “body AI”, basic models built to operate machines. Japan wants to own the software brain. The hardware to build it, however, still comes from Nvidia. The US chip giant is building a “Vera Rubin AI factory”, a large data center full of next-generation chips, expected to be launched in 2028, with 13,750 Vera CPUs and 27,500 Rubin GPUs, delivering 140 megawatts. Noetra will oversee the effort, with plans to build a data center. Noetra’s plan goes through three phases: a thinking model heavy on Japanese language skills starting in fiscal 2026; an omni-modal version that handles text, images, video, and audio in 2028; and “Real-world Native AI” built for robotics in 2030, released to Noetra’s external developers in phases.
A robotics alliance – Japan’s industrial giants are lining up behind Cosmos. Nvidia is targeting a factory floor in Japan, and many of the country’s top robotics and manufacturing players are signing on. Fanuc, Yaskawa, Kawasaki Heavy, Fujitsu, Hitachi, NEC, Sony, SoftBank, Kubota and the robotics group AIRoA say they plan to build on the models of Nvidia’s Cosmos, an open modeling effort Nvidia started in May with several AI labs around the world. In Tokyo, Nvidia gave them reason to commit, unveiling the Cosmos 3 Edge, a version of the model that uses its Jetson Thor chips inside the device itself. Some are already testing a shared control system; others, such as Honda R&D and Omron, are building on tools now. “The next frontier of AI is in the physical world, and this is a once-in-a-generation opportunity for Japan,” Huang said in a company statement. “Japan invented modernity. Now, it has the opportunity to reinvent it in the age of smart industries.”
Toyota — cars and physical AI. Toyota uses Nvidia chips in all of its stacks. It has committed its next-generation cars to Nvidia’s Drive platform at CES in January 2025; The new venture extends Nvidia into manufacturing, where simulation is used to design production lines, into the software that drives its cars, and into systems that learn traffic. Toyota’s cars will use advanced driver assistance, which is steering and braking but still requires a driver, a more thorough approach than Waymo and Tesla, which are building systems that rely less on the human driver.
Huang’s visit put virtual AI at the center of Japan’s industrial strategy, and Tokyo is spending money to support it. Faced with a shrinking workforce, Japan wants 10 million AI-equipped robots in 18 sectors by 2040, backed by $65 billion in public and private AI investment.
The long game is great. Japan’s AI Robotics Strategy, released in March, aims to capture more than 30% of the world’s AI robotics market by 2040, a market Tokyo values at around ¥20 trillion, or about $133 billion. METI is funding a home base model to run the machines, and Noetra’s Nvidia-powered factory is where models of that scale, with trillions of parameters, will be trained. It’s a bet that Japan’s factory floor data and manufacturing base can make for a realistic AI
Under the industrial case there is a private one. As the US and China move forward with big AI, Tokyo wants its own data, its own computing, and less reliance on infrastructure it doesn’t control. Huang appeared on July 16 with trade minister Ryosei Akazawa at the AI’s physical launch, with Prime Minister Sanae Takaichi joining via video. Takaichi’s administration has made AI and semiconductors the centerpiece of a growth plan that is chasing ¥370 trillion ($2.3 trillion) in public and private investment by 2040. Noetra’s factory — which Nvidia bills as “the world’s first national AI infrastructure” — is the clear bet for now. Japan’s push for independence, at least for now, depends on American chips.ndence works on American silicon.
In two days, Huang sat across from almost every important name in Japanese tech – the CEOs of Toyota, Fanuc, Yaskawa, Fujitsu and Kawasaki at lunch, and a bunch of supply chain chiefs over skewers and whiskey at Kanda izakaya.
It’s the same playbook he’s been using for the past few weeks — a key note coming home to Taiwan, fried chicken, and a 50,000-GPU deal in Seoul last fall. This time, it was Tokyo’s turn, with robots, a chain of goods, and chips underneath.
If you shop through links in our articles, we may earn a small commission. This does not affect our editorial independence.
