SoftBank Forms Robotics Company to Build Data Centers, Eyes $100 Billion IPO

SoftBank's Next Big Bet: Building Data Centers with Robots

Amid the explosive growth in demand for AI infrastructure, SoftBank Group is hatching an ambitious plan — establishing a new company dedicated to building data centers using robotics technology, with its eyes set on an IPO valued at up to $100 billion. The news once again underscores Masayoshi Son's determination to go "all in" on AI, while revealing a philosophically compelling industry trend: we need infrastructure to build AI and robots, but we equally need AI and robots to build that infrastructure.

Core Strategy: The "Bootstrapping" Logic of AI Infrastructure

SoftBank's strategic vision is essentially about completing a full loop — using AI-driven robots to accelerate the construction of data centers that house AI computing power. This "building AI with AI" bootstrapping model is becoming a direction that tech giants are racing to explore.

Global data center construction currently faces multiple bottlenecks. On one hand, as demand for large language model training and inference grows exponentially, the pace of data center construction falls far short of the surging demand for computing power. On the other hand, the traditional construction industry faces severe labor shortages, with long construction cycles and soaring costs becoming key constraints on AI industry expansion.

SoftBank's solution strikes at the heart of these pain points: by deploying automated robots in the data center construction process, it aims to dramatically improve construction efficiency, shorten timelines, and reduce labor costs. If this model can be scaled successfully, it would fundamentally transform how AI infrastructure is built.

What Justifies a $100 Billion IPO

The $100 billion IPO target is eye-catching, but given the current scale and growth rate of the AI infrastructure market, this valuation is not without basis.

First, the global data center market is in an unprecedented expansion phase. According to multiple research firms, global data center capital expenditure is projected to reach trillions of dollars over the next five years. Tech giants including Microsoft, Google, Amazon, and Meta are already investing tens of billions of dollars annually in data centers, and this trend is accelerating.

Second, SoftBank's moves in the AI sector have grown increasingly aggressive in recent years. From full ownership of ARM, to massive investments in OpenAI, to deep involvement in the previously announced Stargate project, Masayoshi Son is building a comprehensive AI ecosystem spanning chips, models, and infrastructure. This new robotics data center company is a critical piece of that puzzle.

Third, the sector combining robotics with construction itself holds enormous market potential. If proven successful in data center construction, this model could readily expand to broader industrial and commercial building scenarios.

Industry Landscape and Competitive Dynamics

SoftBank is not the only player to recognize this opportunity. Competition around AI infrastructure has already entered a white-hot phase:

• NVIDIA continues to dominate the GPU computing supply chain, with its data center business revenue hitting record highs
• Microsoft and OpenAI are jointly advancing the Stargate super data center initiative
• Tesla's Optimus humanoid robot is also exploring industrial applications
• Numerous construction tech startups are bringing automation technologies to job sites

SoftBank's differentiating advantage lies in its integration capabilities. Leveraging ARM's chip ecosystem, its investment network spanning multiple AI companies, and Masayoshi Son's personal clout in capital markets, SoftBank is well-positioned to deeply bind robotic construction with AI infrastructure demand, creating a unique competitive moat.

Risks and Challenges

Of course, this grand blueprint also faces significant challenges. The actual performance of robots in complex construction environments remains to be proven, and a massive engineering gap exists between concept and scaled production. Data center construction involves coordinating multiple complex systems including power, cooling, and networking — full reliance on robotic construction is unrealistic in the near term.

Moreover, whether a $100 billion IPO valuation can win investor confidence in the current market environment remains uncertain. SoftBank's previous investment track record with the Vision Fund has raised questions about its judgment, and the new company will need to prove its value with tangible results.

Outlook: An "Infrastructure Revolution" for the AI Industry

Regardless of whether this new company ultimately achieves its $100 billion valuation target, SoftBank's move sends an important signal: competition in the AI industry is extending comprehensively from the "software layer" to the "physical layer." Whoever can build AI infrastructure faster and more efficiently will seize the advantage in the next round of competition.

The concept of "using AI to build AI infrastructure" may well become one of the most important narratives in the tech industry over the next decade. As robots begin constructing the very data centers that house their own intelligence, we are witnessing the dawn of a new era of technological self-evolution.