A major new initiative has just launched to build Japan's very own large-scale AI model.
Led by SoftBank, this new venture, named "Japan AI Base Model Development," brings together a powerhouse team including Sony, Honda, NEC, and Japan's three largest banks. Their goal is ambitious: to create a massive, one-trillion-parameter AI model by 2030, with a special focus on connecting AI to real-world machines and robots, often called 'physical AI'. This isn't just a corporate project; it's a national strategic move to ensure Japan doesn't fall behind in the global AI race and can protect its own data.
But what makes this announcement so significant is that it’s not just an idea on paper. The groundwork has been laid for months, making this plan highly achievable. Let's look at the key factors that led to this moment.
First, there's strong government backing. The Japanese government, through its Ministry of Economy, Trade and Industry (METI), has already committed over 10 trillion yen to support AI and semiconductor development through 2030. A significant portion, around 1 trillion yen, is specifically earmarked for developing domestic 'foundation models'. This new company is perfectly positioned to receive that funding, which greatly reduces the financial risk.
Second, the physical infrastructure is ready. The project will be hosted at a massive AI data-center campus in Sakai. One part of this campus, run by KDDI, is already operational with the latest NVIDIA AI chips. SoftBank also purchased the adjacent facility from Sharp to expand it. This means they don't have to wait years to build data centers from scratch. They can start training their AI model almost immediately, while also ensuring sensitive Japanese data stays within the country, a concept known as 'data residency'.
Third, Japan already has a growing base of AI talent and technology. Companies like NEC and NTT have been developing their own Japanese-language AI models for a while. SoftBank has also been working on its own specialized AI. This existing expertise provides a solid foundation of talent and tools to build upon for this much larger, national-level project.
Finally, there's a clear path to market. Just before this announcement, Microsoft pledged a massive investment in Japan. Crucially, Microsoft's cloud service, Azure, will allow customers to use Japan's sovereign AI infrastructure. This move creates a ready-made customer base for the new AI model, as Japanese companies can use a world-class cloud platform while still keeping their data securely in Japan. Together, these pieces—funding, infrastructure, talent, and market access—turn an ambitious vision into a concrete, executable plan.
- Foundation Model: A large, powerful AI model trained on a vast amount of data. It can be adapted for many different tasks, serving as a "foundation" for more specialized AI applications.
- Physical AI: A field of AI focused on enabling robots and other machines to interact with and understand the physical world, moving beyond just processing digital information.
- Data Residency: The requirement that data be stored on servers located within the borders of a specific country, often for legal, security, or privacy reasons.