日本語 
SoftBank Group is putting another big bet on artificial intelligence. This time, the focus is not a startup, a model, or a software platform. It’s infrastructure.
The company has announced plans to develop a 1GW AI-focused data center campus in northern France through a partnership with French AI infrastructure company Sesterce. The project was unveiled as part of the Choose France 2026 summit and is expected to become one of the largest AI data center developments in Europe.
The number itself stands out. One gigawatt is not a small project. Facilities of that size are built with the expectation that demand will continue growing for years.
That tells you something about where the AI market is heading.
AI Needs More Than Models
For the last couple of years, most conversations around artificial intelligence have focused on what AI can do.
New models. New chatbots. New agents. New enterprise applications.
What often gets overlooked is what sits underneath all of that.
Every AI system needs computing power. A lot of it.
Training large models requires thousands of GPUs running for extended periods. Even after training is complete, serving AI applications to millions of users requires enormous infrastructure. Data centers, networking equipment, cooling systems, power supplies. None of it is cheap.
The problem is that demand is growing faster than infrastructure in many regions.
That’s one reason companies like SoftBank are moving aggressively into this space.
Building AI infrastructure today is starting to look a lot like building railroads during the industrial era. The companies controlling the infrastructure may end up holding some of the strongest positions in the market.
Also Read: LINE Yahoo Subsidiary Deploys AI Agent to Transform QA, Cutting Testing Time by Half
Why France?
At first glance, a Japanese company building a major AI facility in France might seem unusual.
But Europe is becoming increasingly important in the AI conversation.
Governments across the region want more domestic computing capacity. Regulators want greater control over where data is processed. Enterprises want alternatives to relying entirely on infrastructure located elsewhere.
Large-scale facilities like the one planned by SoftBank and Sesterce help address those concerns.
France has also been actively positioning itself as a destination for technology investment. AI infrastructure is now part of that strategy.
The project fits neatly into broader European efforts to strengthen local AI capabilities while attracting international investment.
What This Means for Japan
Even though the facility will be built thousands of kilometers away, Japanese companies are paying attention.
The announcement highlights something many businesses in Japan are already starting to realize. AI competitiveness isn’t just about software anymore.
Having access to computing power is becoming a strategic issue.
Japan has strong technology companies. It has world-class manufacturers. It has expertise in electronics, telecommunications, robotics, and industrial systems.
But the AI era is creating a different set of requirements.
Companies need infrastructure.
They need access to GPUs.
They need data centers capable of supporting large-scale AI workloads.
They need energy.
A few years ago, discussions about AI were mostly centered around algorithms and applications. Now infrastructure is becoming part of the conversation. In some cases, it is becoming the conversation.
The Ripple Effect Across Industries
Projects like this rarely affect just one sector.
Data centers create demand across an entire ecosystem.
Cloud providers need more capacity.
Networking vendors supply equipment.
Cybersecurity firms provide protection.
Cooling technology providers get involved.
Power management companies see new opportunities.
Construction firms, engineering specialists, and facility operators all become part of the picture.
The impact spreads much further than people often assume.
For businesses using AI, additional infrastructure can also help reduce some of the bottlenecks that have emerged as adoption accelerates. Many organizations want to deploy larger AI workloads but are competing for the same computing resources.
More capacity doesn’t solve everything overnight, but it does help.
A New Phase of the AI Race
What’s interesting about SoftBank’s announcement is what it says about the industry’s priorities.
A year or two ago, headlines were dominated by model launches.
Today, some of the biggest announcements involve chips, power generation, data centers, and infrastructure investments.
That’s not happening by accident.
Companies are realizing that advanced AI systems cannot exist without the physical infrastructure supporting them.
The race is expanding beyond software.
It’s becoming a competition over who can build enough capacity to power the next generation of AI services.
SoftBank’s planned facility in France is another sign of that shift.
For Japan’s technology sector, the message is fairly clear. The future of AI will not be determined solely by who develops the smartest models. It will also be shaped by who owns, operates, and controls the infrastructure those models depend on.
And right now, that infrastructure is becoming one of the most valuable assets in the entire technology industry.
English