Masayoshi Son: The Samurai Investor Betting the Farm on AI

In early June 2026, SoftBank Group quietly achieved a historic milestone: it overtook Toyota to become Japan’s most valuable publicly traded company by market capitalization for the first time in more than two decades. Shares surged as the market priced in the global AI boom and SoftBank’s strategic positioning within it.

At the center of this story stands Masayoshi Son — one of the boldest, most polarizing, and arguably strangest successful investors in modern history. His track record is a rollercoaster of jaw-dropping wins and spectacular, headline-grabbing failures. Yet through it all, the SoftBank founder keeps swinging for the fences with outsized conviction.

---

The AI Infrastructure Play

While SoftBank’s stake in OpenAI (and the latter’s rumored IPO path) gets plenty of attention, a deeper and potentially more transformative bet is underway: building the physical backbone of the AI era.

SoftBank recently committed up to €75 billion to develop and operate 5 gigawatts of AI data center capacity across France — one of the largest such projects in Europe. The initial phase targets 3.1 GW by 2031, with sites including Dunkirk. SoftBank will handle construction and network deployment, while partnering with Schneider Electric for the critical power and electronics infrastructure.

This isn’t just capex for its own sake. It aligns with Europe’s growing desire for greater technological sovereignty and reduced reliance on U.S.-centric cloud and AI infrastructure.

Even more ambitious is SoftBank’s robotics gambit. The company is consolidating its various robotics and AI-related assets (including stakes tied to ABB Robotics and others) into a new entity called Roze (or Roze AI).

The goal: deploy fleets of autonomous robots to dramatically speed up the construction of data centers and AI infrastructure.

Why? Because the biggest bottleneck for the AI buildout isn’t money or chips — it’s skilled labor and construction speed. Son’s thesis is that robotics can solve the “hands shortage.” The plan is reportedly to take Roze public in the U.S. as early as late 2026, targeting a valuation around $100 billion. It’s an audacious, vertically integrated bet: invest in the data centers *and* the robots that build them.

---

The Man Behind the Bets

Masayoshi Son has never been a conventional investor. He is famous for making enormous, concentrated bets based on visionary founders and transformative technologies rather than spreadsheets or safe diversification.

Legendary wins include:

• His $20 million investment in Alibaba in 2000 (when the company was virtually unknown outside China). That stake eventually grew into tens of billions of dollars — one of the greatest venture returns in history.
• An early investment in Yahoo, which reportedly turned a modest sum into billions.
• Significant gains from ARM, where SoftBank multiplied its investment before later strategic moves.

These successes gave Son the capital and credibility to launch the massive Vision Fund, which deployed tens of billions into high-growth tech companies.

Spectacular misses are equally well-documented:

• SoftBank poured nearly $20 billion into WeWork. When the company collapsed, it inflicted massive losses and temporarily damaged SoftBank’s reputation and standing.
• Son sold a large Nvidia stake in 2019 for around $3.3 billion in profit — only to watch the value of those shares explode to well over $150 billion in subsequent years amid the AI boom.
• Investments in troubled or failed entities, including exposure to FTX, Greensill Capital, and DiDi (timed poorly ahead of Chinese regulatory action).

Critics have called out the pattern of over-optimism, rapid capital deployment, and occasional poor timing. Supporters point out that the scale of the wins has more than offset the losses — and that Son’s willingness to take outsized risks is precisely what allows for outlier outcomes.

Also read:

• Chinese Open-Weight Models Are Nipping at the Heels of Western SOTA
• The Enterprise AI Gold Rush Is Over — What Comes Next Is Execution
• Sensor Tower’s State of AI 2026 Report: ChatGPT Hits 1 Billion MAU in Record Time, But AI Is Quietly Rewiring E-Commerce
• Google Gives Intel Foundry Its Biggest Chance in Years

---

A Distinctive Style

Son’s approach is often described as founder-focused and vision-driven. He has spoken about investing in people with “burning eyes” and ambitious dreams rather than polished business plans. This philosophy powered his early Alibaba bet and continues to shape SoftBank’s AI-era strategy.

His style is high-conviction and high-leverage. When he believes in something, he goes big — sometimes too big. When things go wrong, the fallout can be dramatic (as with WeWork). Yet he has shown remarkable resilience, continuing to pursue even larger visions.

The current chapter — massive European data center investments combined with a robotics company aimed at solving AI infrastructure bottlenecks — fits the pattern perfectly.

It’s a grand, integrated bet on the physical layer of artificial intelligence, executed with the same swagger that has defined Son’s career.

Whether Roze AI delivers on its $100 billion ambitions or the French data center push accelerates Europe’s AI independence remains to be seen. What’s certain is that Masayoshi Son will remain one of the most fascinating figures to watch in global tech and investing. His bets are rarely boring — they tend to be either legendary successes or memorable disasters.

In a world of cautious capital allocation, Son operates like a modern samurai: decisive, all-in, and unafraid of dramatic swings. For better or worse, that approach has helped make SoftBank one of the most valuable companies in Japan — and keeps the world guessing what comes next.