Japan has committed more than 10 trillion yen in public support for artificial intelligence and semiconductor infrastructure, betting homegrown technology can loosen China's grip on its domestic market. The Diplomat reported the details in an article by Bonnie Girard dated October 5, 2026.
The pitch is simple. Japan trails the United States and China in core AI. It wants to build its own capacity rather than depend on foreign systems that carry foreign risks.
The Price Problem
Dr. Takahiro Tsuchiya of Kyoto University put the market reality bluntly. Chinese companies dominate Japan's domestic market in part because their AI products cost less than Japanese companies' offerings. Price, not politics, is doing much of the work. Tsuchiya said Japan hadn't kept pace with the United States and China in core AI.
He spoke to reporters at the Japanese Embassy in Washington, D.C., and to people attending his lecture at the Japanese Consulate in Boston. He "strongly advocates for a robust U.S.-Japan alliance to counter regional threats," which the article identified as China and North Korea.
But Tsuchiya didn't call for simply following Washington's lead. "While Japan must tighten technology export controls alongside the U.S. to counter China's military-civil fusion, Tokyo's defensive and economic calculations must remain grounded in its own reality," he said. Japan can't copy the United States, he argued, and must make its own assessments of threats and responses. The geography explains why. Japan sits close to China and is deeply connected to it through trade.
Government Money, Private Partners
Tokyo launched a $6.34 billion initiative beginning in fiscal year 2026, bringing Japan's industrial robotics industry together with next-generation AI. The plans cover heavy industry, transport and infrastructure. One example: autonomous AI systems managing Tokyo's heating and cooling infrastructure.
Demographics add pressure. Japan's aging population and resulting labor shortage have led some companies to modernize and build AI capabilities of their own.
The private sector is moving too. Nikkei reported on September 9 that trading house Sumitomo Corp. "has partnered with Sakana AI to bring the startup's artificial-intelligence models to its 100,000 or so client companies." The deal helps the Tokyo-based company sell its solutions to small businesses. Nikkei said Sakana AI's models are "said to be comparable to cutting-edge overseas models."
The partnership speaks to a practical worry. Japanese companies have been reluctant to expose confidential data to generative AI systems. Nikkei said they'd fallen behind U.S. and Chinese rivals in using AI for structural and operational reforms, and that there were "high expectations for Sakana AI as a provider of AI models that do not rely on overseas companies' services." The venture is a Japanese domestic project, with no U.S. or Chinese companies involved.
A Fast Policy Build-Out
Japan moved quickly to establish its AI policy framework. It brought an AI Act into force in June 2025, adopted a formal AI Basic Plan in December 2025, established a Prime Minister-led AI Strategic Headquarters and committed more than 10 trillion yen in public support, all in under a year, according to Hello World Japan.
The Cabinet then passed its Second Artificial Intelligence Basic Plan on July 14, targeting agentic AI and technological sovereignty. A road map aims to mobilize 370 trillion yen ($2.3 trillion) in public and private funding by 2040, fourteen years from now.
The initial plan was candid about the shortfall. It acknowledged "gaps between Japan and other growing countries in utilization rates and private investment," which had "widened" through 2025. Hello World Japan characterized the strategy as "recovery, not regulation," and said: "That is a different story from the 'Japan is cautious on AI' read that still circulates outside the country."
Why This Matters:
Japan's approach shows a government trying to close a competitive gap with a mix of public funding and private-sector partnerships, and the numbers are large. A 370 trillion yen target depends heavily on private capital showing up, and the government's own plan admits private investment has lagged. The Chinese price advantage Tsuchiya described won't disappear just because Tokyo wants it to; Japanese products will have to compete on cost and quality, not just on origin. The security case is clearer. Firms unwilling to hand confidential data to foreign AI systems have a real reason to want domestic options, and the Sumitomo-Sakana deal is a market-driven answer to that demand. Whether the state's spending sharpens that edge or crowds out private effort is the question to watch.