TOKYO: Microsoft announced a $10 billion investment in Japan spanning 2026 to 2029, the largest single-country commitment the company has made in the Japanese market. Vice Chair and President Brad Smith made the announcement during a visit to Tokyo on April 3, framing the deal around three pillars: technology infrastructure, cybersecurity partnerships, and workforce training at a scale Japan has not seen from a single foreign company.
The deal builds on a wave of massive AI spending across the tech industry. Microsoft itself invested $2.9 billion in Japan in April 2024. Two years later, it is more than tripling that figure.
Why Microsoft Is Building AI Infrastructure Inside Japan
The technology pillar of the deal centers on keeping AI compute and data physically inside Japan. Microsoft will collaborate with Sakura Internet and SoftBank to deploy GPU-based AI infrastructure accessible through Azure, with strict data residency. Japanese companies and government agencies will be able to run AI workloads on domestic hardware rather than routing data through overseas data centers.
This matters because Japan’s government has increasingly pushed for what policymakers call “sovereign AI,” where sensitive data processed by AI systems never leaves the country’s borders. GitHub Enterprise Cloud will now offer data residency in Japan for the first time, letting companies with strict governance requirements store code and repository data domestically. Microsoft is also expanding Azure Local, its on-premises cloud product, to support disconnected environments where internet access is unreliable or restricted.
Sakura Internet, a mid-cap Japanese cloud provider, saw its stock jump 20.2% on the announcement. SoftBank Corp. rose 1.02%. The market reaction reflected how much domestic companies stand to gain from being selected as Microsoft’s infrastructure partners rather than competing against its data centers.
On the cybersecurity front, Microsoft is deepening partnerships with Japan’s National Cybersecurity Office and the National Police Agency. The company’s Digital Crime Unit will expand operations targeting transnational scam networks that have increasingly targeted Japanese consumers and businesses.
What Does Microsoft’s $10 Billion Japan Bet Mean for the Global AI Race?
Microsoft’s investment signals that the AI infrastructure buildout has gone global. U.S. data center capacity is stretched thin, and the biggest cloud providers are now competing to lock in overseas markets where demand for AI compute is growing faster than local supply can keep up. Japan, with 94% of Nikkei 225 companies already using Microsoft 365 Copilot, is a market where enterprise AI adoption is already running ahead of infrastructure.
Japan is also staring at a demographic wall. The country’s Ministry of Economy projects a shortfall of 3.26 million AI and robotics workers by 2040, according to government workforce projections. The country will need 4.98 million workers specialized in these fields, but at the current pace of training, only 1.72 million will be available. That gap explains why Microsoft’s pledge to train 1 million people is not corporate philanthropy. It is a market-creation strategy: train the workers, and they build on your platform.
Microsoft says it has already trained 3.4 million people in Japan in AI skills over the past two years, exceeding its original 2024 target of 3 million. The next phase will focus on enterprise-grade skills through partnerships with Fujitsu, Hitachi, NEC, NTT Data, and SoftBank. Separately, the Japanese Electrical Electronic and Information Union will provide foundational AI training to 580,000 workers. One in five working-age Japanese already uses generative AI, above the global average of one in six.
For founders building products that depend on cloud AI, the infrastructure race reshaping the chip and data center industry directly affects pricing and availability. More compute capacity in more countries means more competitive pricing and lower latency for companies serving Asian markets.
What’s Next
Microsoft also announced a $1 million research grant program focused on large-scale AI analysis and a fellowship program for research leaders in Japan. Brad Smith said in the company’s press release that the investment “will enable us to meet the country’s growing demand for cloud and AI services.”
The next milestone to watch is whether SoftBank and Sakura Internet can actually build out GPU capacity fast enough to absorb Microsoft’s investment within the four-year window. Sakura Internet is a fraction of the size of hyperscalers like AWS or Google Cloud, and scaling GPU infrastructure at the pace Microsoft is promising will test the company’s operational capacity. If it works, the SoftBank-Sakura-Microsoft model could become a template for how Big Tech enters markets where AI capital is flooding in but local infrastructure has not caught up.
Japan’s government has signaled it will keep pushing for domestic AI capacity. The country committed ¥1.23 trillion to AI in its latest budget cycle, and regulators are expected to tighten data residency requirements further in 2027. For Microsoft, getting $10 billion of infrastructure into the ground before those rules harden is as much about regulatory positioning as it is about revenue.
