SoftBank: The Internet Conglomerate's Wild Ride

I. Introduction & Episode Roadmap

Picture this: A Korean-Japanese entrepreneur walks into a meeting with Steve Jobs in 2005, pulls out a sketch of an iPod with a phone dial pad crudely drawn on it, and says, "Steve, you should make this." Jobs laughs him off. Three years later, that same entrepreneur secures exclusive iPhone distribution rights for Japan, fundamentally reshaping the country's mobile market. That entrepreneur? Masayoshi Son, founder of SoftBank—a company that has redefined what it means to be an investor in the digital age.

How does a company go from distributing PC software in 1980s Japan to writing $100 billion checks that make Silicon Valley VCs look like they're playing with pocket change? How does the same firm that turned $20 million into $60 billion with Alibaba also lose $13 billion on WeWork? And perhaps most intriguingly, how has one man's almost religious belief in the "information revolution" created a corporate entity that defies every traditional category—part telecom operator, part semiconductor owner, part venture capitalist, part prophet of technological singularity?

SoftBank isn't just a company; it's a financial phenomenon that has fundamentally altered how technology gets funded globally. Ranked 461st in the 2024 Forbes Global 2000, with a market cap that has swung between $30 billion and $200 billion over the past decade, SoftBank represents something unprecedented: a Japanese conglomerate that became Silicon Valley's biggest patron, a traditional corporation that operates like a hedge fund, and a bet-the-company culture that makes even the most aggressive American entrepreneurs look conservative.

This is the story of three acts in SoftBank's evolution. Act One: The scrappy software distributor that saw the internet coming before almost anyone else. Act Two: The telecom-turned-investment hybrid that made one of venture capital's greatest trades with Alibaba. Act Three: The Vision Fund era, where SoftBank didn't just participate in the venture capital game—it fundamentally rewrote its rules, for better and worse.

What follows is not just a corporate history but a meditation on ambition, timing, and the thin line between visionary and reckless. It's about understanding how Masayoshi Son's "300-year vision" has created both breathtaking successes and spectacular failures, often simultaneously. And it's about what happens when you give an entrepreneur with the risk tolerance of a BASE jumper access to more capital than most countries' sovereign wealth funds.

↑ Back to Top

II. Origins & The Visionary Founder (1981–1990s)

The moment that changed everything came in 1980, when a 23-year-old Masayoshi Son picked up a copy of Popular Electronics at UC Berkeley. The cover featured a photograph of a microchip—just a tiny square of silicon with intricate patterns. To most people, it was technical ephemera. To Son, it was a glimpse of the future. He stared at that image and saw not just a component, but the foundation of an entirely new civilization. "The tears rolled down my face," he would later recall. "I knew then that the microchip would change humanity."

This wasn't hyperbole—it was the founding mythology of what would become SoftBank. Son had already shown entrepreneurial flair, having sold a pocket translator invention to Sharp for $1.7 million while still in college. But the microchip revelation was different. It wasn't about a single product; it was about positioning himself at the center of a technological revolution he was certain would transform society.

On September 3, 1981, Son founded Nihon SoftBank in Tokyo with just two part-time employees and a vision that seemed absurd to everyone except him. The company name itself was deliberate: a "bank" of software that would serve as infrastructure for Japan's information society. While Japan's corporate titans were perfecting hardware manufacturing, Son saw that software would be the real battleground. His timing was either prescient or lucky—probably both. Personal computers were just beginning their march from hobbyist curiosity to business necessity.

The early SoftBank was nothing like the financial colossus it would become. Son started by distributing packaged PC software, essentially serving as a middleman between American software companies and Japanese retailers. His first major breakthrough came from exploiting an information asymmetry that seems quaint today: Japanese businesses desperately needed PC software but had no idea how to source it from American developers. Son positioned SoftBank as the bridge.

By 1982, he had secured exclusive distribution rights for several American software titles in Japan. The business model was elegant in its simplicity—buy software licenses in bulk from U.S. companies eager to crack the Japanese market, then distribute them through a network Son painstakingly built from scratch. Within a year, SoftBank controlled 50% of Japan's PC software distribution market. This wasn't through superior technology or innovation; it was pure commercial execution and relationship building.

The expansion into publishing came almost accidentally. Son realized that Japanese consumers didn't just need software—they needed to understand it. In 1982, he launched Oh!PC and Oh!MZ, magazines dedicated to personal computing. These weren't just publications; they were evangelism tools for the PC revolution. By 1990, SoftBank's publishing arm, SoftBank Publishing, was producing over 100 magazine titles and generating more profit than the software distribution business.

The trade show business followed similar logic. If you're already the nexus of Japan's PC industry, why not bring everyone together physically? SoftBank began organizing computer trade shows that became must-attend events for anyone in Japanese tech. Each new business line reinforced the others—magazines promoted the trade shows, trade shows generated software distribution deals, and software distribution provided content for the magazines.

What distinguished Son from his contemporaries wasn't just ambition but his particular brand of it. While other Japanese entrepreneurs thought in terms of incremental improvement and consensus building, Son operated with an almost American swagger mixed with Korean-Japanese outsider hunger. He would regularly announce goals that seemed delusional—"We'll be a $100 billion company in 20 years"—then work backwards from that vision to figure out how to make it happen.

The cultural challenges were immense. As a Korean-Japanese entrepreneur in a society that prized ethnic homogeneity, Son faced discrimination that would have deterred most. Banks wouldn't lend to him. Established companies viewed him with suspicion. The Japanese business establishment, with its emphasis on lifetime employment and gradual promotion, saw Son's aggressive American-style capitalism as almost offensive. He responded by being even more aggressive, even more American.

By 1990, SoftBank had grown from two employees to over 700, with revenues approaching $400 million. But Son was just getting started. He studied historical patterns obsessively, particularly the rise of American robber barons like Rockefeller and Carnegie. His conclusion: every major economic transformation created opportunities for new empires, but only for those willing to move fast and bet big. The PC revolution was just the first wave. The real opportunity would come with whatever came next.

The April 1994 IPO on the Tokyo Stock Exchange marked SoftBank's transformation from successful private company to public market player. The offering raised $140 million, giving Son the war chest he needed for his next moves. But more importantly, it gave him currency—shares he could use for acquisitions. The IPO prospectus laid out a vision that seemed fantastical at the time: SoftBank would become the world's largest provider of digital infrastructure. Investors who bought in were betting not on what SoftBank was, but on Son's ability to see around corners.

The Ziff-Davis acquisition in February 1996 proved Son could play on the global stage. After failing to acquire the publishing giant in 1994, Son came back with a $2.1 billion offer that stunned both American and Japanese business communities. Ziff-Davis published PC Magazine, PC Week, and MacWEEK—crown jewels of tech media. The price seemed insane for a publishing company, but Son wasn't buying magazines. He was buying information flow and influence in the emerging digital economy.

Critics called him reckless. The Japanese press wondered why a Japanese company was spending billions on American magazines when the internet might make print obsolete. Son's response revealed his true strategy: "We're not buying their past; we're buying their future digital transformation." He immediately pushed Ziff-Davis online, creating ZDNet, which became one of the web's first major tech news destinations.

By the mid-1990s, SoftBank had evolved from software distributor to something harder to categorize—part publisher, part event organizer, part technology investor, all held together by Son's conviction that information would be the oil of the 21st century. The company's seemingly scattered portfolio actually followed a coherent logic: own the infrastructure of the information revolution, whether that meant software distribution, publishing, or whatever came next.

What came next would transform SoftBank from a successful Japanese company into a global financial force. Son had been tracking a small startup called Yahoo! and an even smaller Chinese entrepreneur named Jack Ma. The internet age was about to begin, and Son was determined to own as much of it as possible.

↑ Back to Top

III. The Internet Revolution & Yahoo! Japan (1995–2000)

In November 1995, Masayoshi Son boarded a plane to California with a plan that his own executives thought was insane. He was going to meet Jerry Yang and David Filo, the founders of a year-old startup called Yahoo! that was basically just a directory of websites. Son didn't want to just invest—he wanted to bring Yahoo! to Japan. His proposal was audacious: he would put up most of the capital, do all the work, but Yahoo! would maintain significant ownership. His own team called it the worst deal they'd ever seen. Son called it the beginning of the Japanese internet.

The meeting in Yahoo!'s cramped Sunnyvale offices lasted less than two hours. Son sketched out his vision on a whiteboard: Japan had minimal internet penetration, but it had high disposable income, density that made infrastructure buildout economical, and a culture obsessed with new technology once properly introduced. Yahoo! Japan wouldn't just be a translation—it would be a complete reimagining of the internet for Japanese users. Yang and Filo, still adjusting to their sudden success, were bewildered by this intense Japanese executive who seemed to understand the internet's potential better than most Americans.

The deal structure was vintage Son—seemingly irrational but actually brilliant. SoftBank would own 60% of Yahoo! Japan, Yahoo! would own 35%, and the remaining 5% would go to other investors. SoftBank would fund everything and do all the work. Yahoo! would provide the brand and technology. Critics pointed out that Son was essentially paying to build someone else's brand in Japan. What they missed was that Son didn't care about owning the brand—he cared about owning the market.

Yahoo! Japan launched in April 1996, just as Japan's internet population was exploding from 3 million to 10 million users in a single year. But Son's masterstroke wasn't the timing—it was the localization. While competitors like Lycos and Excite simply translated their American sites, Yahoo! Japan rebuilt everything from scratch. The homepage was denser, with more links—catering to Japanese users' preference for information density. The categories were completely reorganized around Japanese interests: baseball got prime placement, as did anime and manga sections that didn't exist on Yahoo.com.

The company also pioneered something that would become central to Japanese internet culture: anonymous message boards. While American sites pushed real-name policies, Yahoo! Japan embraced pseudonymity, understanding that Japanese users valued privacy and indirect communication. The message boards exploded in popularity, becoming Japan's equivalent of Reddit a decade before Reddit existed.

By 1997, Yahoo! Japan had become the country's homepage. But Son was already thinking three moves ahead. He realized that unlike in America, where internet access came through existing phone lines, Japan needed new infrastructure. So SoftBank began building it, launching Yahoo! BB in 2001—a broadband service that would fundamentally reshape Japanese internet access. The service offered 8 Mbps speeds for $22 per month when NTT was charging $60 for 1.5 Mbps. It was predatory pricing that lost money on every customer, but Son didn't care about short-term profits. He was building a platform.

The dot-com era transformed SoftBank from a successful Japanese company into something unprecedented. Between 1995 and 2000, Son invested in over 800 internet companies, deploying capital with a speed and scale that made him Silicon Valley's most important foreign investor. E*Trade, GeoCities, Buy.com, Webvan—if it ended in ".com," Son probably owned a piece of it. His investment philosophy was simple to the point of recklessness: the internet would be bigger than anyone imagined, so own as much of it as possible.

The numbers from this period are staggering. SoftBank's market capitalization reached $200 billion in February 2000, briefly making Son richer than Bill Gates with a paper net worth of $78 billion. For three days, a Korean-Japanese entrepreneur who had started with nothing was the world's richest person. The Japanese media went into overdrive—here was validation that Japan could compete in the new economy, that it wasn't just about manufacturing anymore.

At the peak, SoftBank's portfolio of internet investments was valued at over $100 billion. Son gave speeches about the "internet lifestyle" and predicted that e-commerce would reach $10 trillion by 2005 (it wouldn't hit that number until 2017, but his direction was correct). He became a celebrity in Japan, his books became bestsellers, and young entrepreneurs began copying his aggressive style—a marked departure from traditional Japanese business culture.

The infrastructure Son built during this period would outlast the bubble. Yahoo! Japan didn't just survive the dot-com crash—it thrived, becoming one of the few internet companies to maintain dominance in its home market against American competitors. While Google conquered most of the world, Yahoo! Japan held firm, controlling over 50% of Japanese search well into the 2010s. The broadband infrastructure SoftBank built became the foundation for Japan having some of the world's fastest internet speeds.

But the real prize from this era wasn't Yahoo! Japan or the broadband network. In 1999, during a trip to China, Son met a former English teacher named Jack Ma who was trying to build a business-to-business marketplace called Alibaba. Ma's presentation was terrible—his English was broken, his slides were amateurish, and his business model was unclear. But Son saw something others missed: Ma understood the Chinese market the way Son understood Japan. The meeting lasted six minutes. Son decided to invest $20 million.

The dot-com crash of 2000 would wipe out $70 billion of SoftBank's market value in a matter of months. Son's net worth plummeted from $78 billion to $7 billion. Of the 800 internet companies SoftBank had invested in, over 700 would fail. The Japanese press, which had celebrated Son as a visionary months earlier, now called him a gambler who had lost. International investors wrote him off as another casualty of irrational exuberance.

But hidden in the wreckage was that single Chinese investment—$20 million in a company no one had heard of, run by a former teacher who couldn't even present properly. That investment would eventually be worth more than SoftBank's entire market cap at the height of the dot-com bubble. The internet revolution had made Son a billionaire, cost him most of his fortune, and secretly planted the seed for one of the greatest investment returns in history.

↑ Back to Top

IV. The Alibaba Bet: One of VC's Greatest Returns (2000)

The boardroom at Beijing's Kempinski Hotel was sweltering on October 31, 1999. Dozens of Chinese entrepreneurs had gathered for what SoftBank's newly launched China venture fund called a "speed-dating" session—each founder got exactly ten minutes to pitch. Masayoshi Son sat at the head of the table, exhausted from seeing pitch after pitch of derivative business models. Then a slight man with an oversized head and animated gestures walked in. Jack Ma, described as an elfin figure, stood out because he spoke the best English among all the presenters that day.

Ma's presentation was, by any objective measure, terrible. His slides were amateurish PowerPoint templates. His business model—an electronic Yellow Pages connecting Chinese manufacturers with overseas buyers—made little sense as explained. At that time, Alibaba was pre-revenue and didn't have a well-defined business model. The company had burned through its initial $600,000 in funding and was weeks away from insolvency. Ma himself looked nothing like a typical tech entrepreneur—he was a former English teacher from Hangzhou who had discovered the internet just four years earlier during a trip to Seattle.

But something extraordinary happened in that room. Son said in a 2017 Bloomberg interview that he instantly decided to invest in Alibaba after seeing Ma's "strong ... shining eyes", despite the Chinese firm having "no business plan and zero revenue" at that time. "His business model was wrong," Son said. But Son added that he was moved by "the way [Ma] talks, the way he can bring young Chinese people to follow him".

Ma said in a 2008 speech that his first meeting with Son in 2000 lasted only six minutes, but the two "fell in love at first sight". Son was keen to invest US$40 million into Alibaba, but Ma eventually agreed to a US$20 million outlay from SoftBank. This negotiation revealed something crucial about both men. Son wanted to put in more money to own more of the company—classic venture capital thinking. Ma wanted less money to maintain more control—understanding that in China's political environment, foreign ownership could become a liability.

In 2000, Japanese telecom giant SoftBank, led by Masayoshi Son, invested $20 million for a 34% stake in Alibaba. The investment came at the absolute peak of dot-com mania, just weeks before the NASDAQ began its historic collapse. SoftBank's own stock was about to lose 99% of its value. Of the 800 internet companies Son had invested in, this one seemed no more likely to succeed than the others. It was, to use venture capital terminology, a spray-and-pray investment made during the final moments of a bubble.

What Son saw that others missed wasn't Alibaba's business model—even he admitted it was wrong. It was the macro opportunity combined with the micro execution. From 1999 to 2000, the number of internet users in China more than doubled, creating a vast market potential for e-commerce. China was starting from effectively zero in e-commerce, with no established players, no consumer habits to change, and no legacy infrastructure to work around. It was a blank canvas the size of a continent.

But more importantly, Son recognized in Ma something he saw in himself—an outsider's hunger combined with an insider's understanding of local culture. Ma, like Son, faced discrimination (Ma for his unusual appearance and humble background in class-conscious China). Both men understood that in societies where you don't fit in, you have to build your own ecosystem. And both had an almost mystical belief in the internet's power to level playing fields.

The relationship that developed between Son and Ma transcended typical investor-founder dynamics. They became genuine friends, calling each other for advice on everything from corporate strategy to personal matters. Ma would later join SoftBank's board, and Son would join Alibaba's—a cross-pollination that was unusual in venture capital. They shared a philosophy that business wasn't just about making money but about transforming society. This sounds like corporate PR speak, but in their case, it was genuine. Both men really believed they were building infrastructure for a new world.

"When I first met Jack Ma, I knew immediately he had the vision and passion to build the world's leading e-commerce company, and I was very happy to invest alongside him to help him realize his ambition." said SBG Chairman and CEO Masayoshi Son. "This investment has been phenomenally successful and, over the past 16 years, we have built a close relationship, working together on many exciting projects. In that time, we have not sold any Alibaba shares" (Son said this in 2016 when SoftBank made its first partial sale).

Alibaba's growth over the next decade validated every assumption Son had made about China's internet potential. The company expanded from B2B into consumer e-commerce with Taobao, digital payments with Alipay, and cloud computing with Alibaba Cloud. Each new business line reinforced the others, creating the kind of ecosystem effects that American tech companies could only dream of. By 2010, Alibaba had become China's e-commerce infrastructure, processing more transactions than Amazon and eBay combined.

The numbers from the Alibaba investment are staggering enough to bear repeating: On 19 September 2014, Alibaba's American initial public offering (IPO) on the New York Stock Exchange raised US$25 billion, giving the company a market value of US$231 billion and, by far, then the largest IPO in world history. This also significantly increased the value of SoftBank's stake in Alibaba to well over $60 billion. The $20 million investment had become worth more than SoftBank's entire market capitalization.

But the Alibaba investment meant more than just financial returns. It fundamentally changed how Son thought about investing. Traditional venture capital focused on business models, unit economics, and path to profitability. The Alibaba experience taught Son that in technology investing, three things mattered more than everything else combined: massive markets undergoing disruption, founders with messianic conviction, and timing. Get those three right, and the business model details would sort themselves out.

This philosophy would drive every major SoftBank investment decision for the next two decades. When Son met Adam Neumann of WeWork, he saw Ma's charisma. When he looked at Uber, he saw Alibaba's winner-take-all dynamics. When he launched the Vision Fund, he was trying to institutionalize the Alibaba playbook: find companies attacking trillion-dollar markets, give them so much capital they couldn't fail, and own massive stakes in the winners.

The end of the Alibaba saga came gradually, then suddenly. SoftBank, which invested US$20 million into Alibaba when it was still a start-up in 2000, said in a corporate filing on Thursday that it was set to book a gain of 1.26 trillion yen (US$8.5 billion) – about 425 times the value of its initial outlay – for the Tokyo-based firm's 2024 financial year after divesting its shares via subsidiary Skybridge. After 24 years, SoftBank had essentially exited its greatest investment, transforming from Alibaba's largest shareholder to a minority holder.

The Alibaba investment remains one of the greatest venture capital returns in history—a 3,000x return that generated more profit than most venture funds see in their entire existence. But its real legacy was psychological. It convinced Son that he wasn't just an investor but a visionary who could see the future. That confidence would lead to both his greatest triumphs and his most spectacular failures. Every subsequent bet—Sprint, ARM, the Vision Fund, WeWork—was Son trying to recreate that moment in Beijing when he looked into Jack Ma's eyes and saw the future.

↑ Back to Top

V. Telecom Transformation: Vodafone Japan Acquisition (2006)

The meeting at Vodafone's London headquarters in January 2006 was tense. Arun Sarin, Vodafone's CEO, had just delivered the news that would define SoftBank's next decade: Vodafone was ready to sell its failing Japanese operations. For Masayoshi Son, sitting across the polished conference table, this wasn't just an acquisition opportunity—it was the chance to transform SoftBank from an internet investor into a real operating company. The price tag was staggering: ¥1.75 trillion (about $15.1 billion), making it the largest acquisition in Japanese corporate history at the time. Son didn't have the money. But he had something more valuable: conviction that Japan's mobile market was about to explode.

On 17 March 2006, SoftBank announced its agreement to buy Vodafone Japan, giving it a stake in Japan's $78 billion mobile markets. On 17 March 2006, Vodafone Group announced it had agreed to sell Vodafone Japan to SoftBank for about US$15.1 billion. On 18 May 2006, the unit was renamed "SoftBank Mobile Corp.", effective 1 October 2006.

The acquisition was audacious for multiple reasons. SoftBank had never operated a telecom network. It was buying a failing asset—Vodafone Japan had lost 1.5 million subscribers in the previous two years and was bleeding market share to NTT DoCoMo and KDDI. The company's 3G network was inferior, its handset lineup was outdated, and its brand was associated with poor coverage and dropped calls. Industry analysts called it the worst acquisition in Japanese history before Son had even closed the deal.

Instead, Vodafone sold off Japan Telecom bit-by-bit to the SoftBank Group in a large number of transactions, the biggest one the sale of Vodafone KK (= Vodafone Japan) to Softbank on 17 March 2006 for about US$ 15 billion. The acquisition of Vodafone KK (=Vodafone Japan) by SoftBank laid the foundation for SoftBank's meteoric rise to a major global player.

But Son understood something the critics missed: Vodafone had failed not because the Japanese mobile market was weak, but because Vodafone had fundamentally misunderstood it. Japan knowledge at HQ, and knowledge at HQ about the specifics of Japan's telecom sector (or lack thereof). As a consequence of far too low investment budgets, Vodafone failed three times to introduce 3G services in Japan. (3G services were not successfully introduced until after the acquisition by Softbank, and after conversion of Vodafone KK to Softbank-Mobile).

The financing structure revealed Son's financial creativity. He created BB Mobile Corp., a special purpose vehicle, to make the acquisition. Yahoo! Japan came in as a co-investor, providing both capital and the promise of content integration. Son borrowed $11 billion from a consortium of banks, using the future cash flows of the mobile business as collateral—essentially buying Vodafone Japan with its own future earnings. The debt load was crushing, but Son calculated that if he could stop the subscriber bleeding and increase average revenue per user by just 10%, the math would work.

The turnaround started immediately. Within days of the acquisition closing in April 2006, Son moved Vodafone's headquarters from the prestigious Atago Green Hills to SoftBank's offices in Shiodome. Most foreign executives were sent home. The entire product strategy was scrapped. On Friday April 21, 2006, SoftBank announced the decision to increase the investments to YEN 250 billion to increase the number of 3G base stations from 20,000 to 30,000. This was more investment in infrastructure in one year than Vodafone had made in the previous five.

But the real transformation came from understanding Japanese consumer psychology. While Vodafone had tried to import European strategies—focusing on prepaid plans and business users—Son realized Japanese consumers wanted something entirely different. They wanted the latest technology, beautiful design, and seamless integration with their digital lives. They wanted phones that were fashion accessories, gaming devices, and personal computers rolled into one.

The masterstroke came in 2007 when Son flew to Cupertino to meet Steve Jobs. Apple was preparing to launch the iPhone, and Jobs was looking for international partners. Every major Japanese carrier had turned him down, believing the iPhone's lack of features standard in Japanese phones (TV tuners, electronic wallets, infrared communication) made it unsuitable for the market. Son saw it differently. He understood that the iPhone represented a paradigm shift—from feature phones to smartphones, from telecom-controlled services to an open app ecosystem.

On 4 June 2008, SoftBank Mobile announced a partnership with Apple and brought the iPhone (3G) to Japan later in 2008. SoftBank Mobile was the only official carrier of the iPhone in Japan until the release of iPhone 4S in 2011, when au by KDDI began to offer it. As the exclusive provider of iPhone 3G in Japan, the company released the device on July 11, 2008. People queued in front of SoftBank shops throughout Japan.

The iPhone exclusivity transformed SoftBank Mobile from industry joke to innovation leader almost overnight. The company gained 3 million net subscribers in 2008 alone. Market share jumped from 16% to 20%. Average revenue per user increased by 15% as customers migrated to expensive data plans. The same network that had been mocked for poor quality suddenly became the only way to get the world's most desirable phone.

Son's pricing strategy was equally revolutionary. While DoCoMo and KDDI maintained complex pricing structures with hundreds of different plans, SoftBank introduced simple, transparent pricing. The "White Plan" offered free calling between SoftBank users—a radical concept in Japan where carriers charged by the minute. Data plans were unlimited at a time when competitors charged by the megabyte. These weren't sustainable pricing models, but Son didn't care about short-term profitability. He was buying market share.

The integration with Yahoo! Japan created synergies that Vodafone could never have achieved. Yahoo! became the default homepage on SoftBank phones. Yahoo! Mail became the default email service. Yahoo! Auctions, Yahoo! Shopping, and Yahoo! News were pre-installed on every device. This wasn't just cross-promotion—it was ecosystem building. Every SoftBank customer became a Yahoo! user, and every Yahoo! user was incentivized to become a SoftBank customer.

By 2010, SoftBank Mobile had become Japan's fastest-growing carrier. Subscriber count had grown from 15 million to 25 million. The company was adding more net subscribers per month than DoCoMo and KDDI combined. The network that had been Japan's worst was now winning awards for customer satisfaction. Revenue had doubled. EBITDA margins had expanded from 15% to 30%.

But the real value of the Vodafone acquisition wasn't financial—it was strategic. It gave SoftBank a platform business generating predictable cash flows that could fund Son's increasingly ambitious investments. It proved that SoftBank could operate real businesses, not just invest in them. Most importantly, it gave Son credibility when he walked into rooms with telecom executives around the world.

The mobile business also taught Son crucial lessons about platform economics. In telecom, the infrastructure investment is massive upfront, but the marginal cost of adding users is minimal. Network effects are powerful—the more users you have, the more valuable the network becomes. Winner-take-all dynamics apply—in most markets, three players capture 90% of profits. These lessons would inform every major decision Son made over the next decade, from the Sprint acquisition to the Vision Fund's investment strategy.

Through its acquisition of Vodafone, SOFTBANK will become an integrated telecommunications service provider, offering both fixed-line and mobile communications over 26 million service lines, generating annual revenue of approximately 2.5 trillion yen on a consolidated basis.

The transformation of Vodafone Japan into SoftBank Mobile represents one of the greatest turnarounds in telecom history. In five years, Son took a failing asset losing hundreds of thousands of subscribers per quarter and transformed it into Japan's most innovative carrier. The ¥1.75 trillion acquisition price that seemed insane in 2006 looked like a bargain by 2011. The mobile business was generating over ¥500 billion in annual EBITDA, funding Son's global ambitions.

Looking back, the Vodafone acquisition was the moment SoftBank transformed from an investment company that happened to own operating businesses into an operating company that happened to make investments. It set the template for everything that followed: identify a massive market in transition, acquire undervalued assets, invest heavily to gain market share, and use the resulting cash flows to fund the next big bet. The strategy would work brilliantly with Sprint—until it didn't.

↑ Back to Top

VI. Sprint Gamble: The $21B American Adventure (2013-2020)

Dan Hesse, Sprint's CEO, was exhausted. It was October 2012, and he had just finished his third marathon board meeting in a week. The topic was always the same: should Sprint accept SoftBank's takeover offer? The company was hemorrhaging cash, losing customers to Verizon and AT&T, and sitting on spectrum assets it couldn't afford to deploy. When Masayoshi Son walked into Sprint's Overland Park headquarters with an offer to invest $20.1 billion for 70% of the company, it wasn't so much a negotiation as a lifeline. "I will make Sprint the number one carrier in America," Son declared with characteristic bravado. The Sprint board thought he was delusional. They took the money anyway.

On 15 October 2012, SoftBank announced plans to take control of American Sprint Nextel by purchasing a 70% stake for $20 billion. On 6 July 2013, the United States Federal Communications Commission approved SoftBank's acquisition for $22.2 billion for a 78% ownership interest in Sprint. The final structure was even more aggressive than initially planned: In July 2013, SBG invested JPY 2.1 trillion (JPY 0.4 trillion in cash-on-hand and JPY 1.7 trillion in borrowings) to acquire Sprint Corporation ("Sprint") as a subsidiary.

The Sprint acquisition represented Son's most audacious bet yet—attempting to replicate the Vodafone Japan playbook in the world's most competitive telecom market. The parallels were obvious: Sprint, like Vodafone Japan, was the struggling third player in a three-player market. It had valuable spectrum assets but lacked the capital to deploy them. Customer service was terrible, the network was fragmented across multiple incompatible technologies, and the brand was associated with dropped calls and slow data speeds.

But the differences were even more striking. The U.S. market was four times larger than Japan's but also far more competitive. Verizon and AT&T had massive scale advantages and deep government relationships. American consumers, unlike Japanese ones, were price-sensitive and disloyal, switching carriers for a $10 monthly discount. Most critically, Son couldn't play the iPhone card again—every U.S. carrier already had it.

Son's strategy was characteristically bold: outspend everyone on network infrastructure, merge with T-Mobile to create a viable third competitor, and ride the 5G wave to market leadership. The plan required perfect execution, cooperative regulators, and about $40 billion in additional capital. It would get none of these things.

The first warning sign came in 2014. In December 2013, multiple reports indicated that Sprint Corporation and its parent company, SoftBank, were working towards a deal to acquire a majority stake in T-Mobile US for at least US$20 billion. The proposed merger would have further bolstered T-Mobile's position in the overall market and resulted in the country's major national carriers being controlled by only three companies. Members of the government were skeptical that such an acquisition would be approved by regulators, citing antitrust concerns and an explicit goal by FCC chairman Tom Wheeler to maintain four national carriers in the United States.

On August 4, 2014, Bloomberg reported that Sprint had abandoned its bid to acquire T-Mobile, considering the unlikelihood that such a deal would be approved by the U.S. government and its regulators. The Obama administration had made it clear: four national carriers were a regulatory red line. Son had spent $2 billion on advisors and preparation for a deal that would never happen.

Without T-Mobile, Sprint was stuck in no-man's land. It lacked the scale to compete with Verizon and AT&T on network quality. It lacked the marketing savvy to compete with T-Mobile on value. Son poured money into the business—over $35 billion in capital expenditure between 2013 and 2017—but it was like pouring water into a bucket with a hole in the bottom. Sprint's network improved, but Verizon and AT&T improved faster. Customer acquisition costs skyrocketed as Sprint offered increasingly desperate promotions: half-price plans, free service for a year, buy-one-get-four-free deals.

The operational challenges were immense. Sprint's network was a Frankenstein's monster of incompatible technologies: CDMA, WiMAX, LTE on three different spectrum bands. The company was simultaneously trying to shut down old networks, build new ones, and migrate customers between them. Every network decision involved trade-offs: improve coverage in cities or expand rural reach? Deploy mid-band spectrum for capacity or high-band for speed? The answer was always "both," which meant doing neither well.

Marcelo Claure, who replaced Hesse as CEO in 2014, brought Latin American flair and cost-cutting discipline. He slashed expenses, fired thousands of employees, and renegotiated every vendor contract. But cost-cutting couldn't solve Sprint's fundamental problem: it was subscale in a scale business. Every quarter brought the same story: subscriber losses, network investments that didn't move the needle, and promises that next quarter would be better.

The 2016 election changed everything. The Trump administration signaled a more merger-friendly approach to telecom consolidation. By 2017, T-Mobile and Sprint were back at the negotiating table. However, after months of speculation and rumors about a potential deal being reached, both T-Mobile and Sprint announced on November 4, 2017, that while they have had discussions about a possible merger, the two parties had decided to end merger talks due to not being able to agree on the terms of the deal, due to Softbank's board of directors reported vote on October 27 where they decided not to give up control of Sprint. Sprint and T-Mobile once again resumed talks of a merger in April 2018 and announced a merger agreement on April 29.

The merger agreement was complex beyond belief. Under the terms of the transaction, Sprint shareholders will receive a fixed exchange ratio of 0.10256 T-Mobile shares for each Sprint share, or the equivalent of approximately 9.75 Sprint shares for each T-Mobile share. As previously announced, per a separate arrangement, SoftBank Group Corp. surrendered approximately 48.8 million T-Mobile shares acquired in the merger to New T-Mobile immediately following the closing of the transaction, making SoftBank's effective ratio 11.31 Sprint shares per T-Mobile share.

The regulatory approval process was a two-year nightmare. The Department of Justice wanted a fourth carrier to replace Sprint. State attorneys general sued to block the deal. The FCC demanded rural coverage commitments. On June 11, 2019, several news outlets reported that the proposed merger between Sprint and T-Mobile was facing a major legal challenge as ten attorneys general from nine states (New York, California, Connecticut, Colorado, Maryland, Michigan, Mississippi, Virginia, and Wisconsin) and Washington, D.C., filed suit to block the merger, alleging it would result in higher prices for consumers to the extent of $4.5 billion annually. It was then reported on June 13, 2019, that Judge Victor Marrero of the United States District Court for the Southern District of New York (SDNY) had set a pre-trial hearing for the week of June 17, 2019, for the merger. On June 21, 2019, it was reported that four more states — Hawaii, Massachusetts, Minnesota, and Nevada — had joined the lawsuit seeking to block the merger.

The breakthrough came with a Solomonic solution: Dish Network would buy Sprint's prepaid business and enough spectrum to become a fourth carrier. When T-Mobile bid $26 billion to acquire Sprint in 2018, it had to get regulatory approval from the FCC and the Justice Department. Both wanted another carrier to replace Sprint as the nation's fourth largest. A deal was made and T-Mobile sold all of Sprint's pre-paid assets including 9.3 million customers, Boost's retail network, employees, and plans to Dish Network.

SoftBank Group Corp. ("SBG") announced the completion on April 1, 2020 (ET) of the merger of Sprint Corporation (a U.S. subsidiary of SBG; "Sprint") and T-Mobile US, Inc. ("T-Mobile") in an all-stock transaction. After seven years and tens of billions in losses, Son's Sprint adventure was over.

The financial engineering that followed was vintage Son—turning a failed acquisition into a profitable trade through complex derivatives. This action follows the satisfaction on December 22, 2023 (U.S. time) of the condition set forth in the Letter Agreement, dated as of February 20, 2020 (the "Letter Agreement"), by and among T-Mobile, SBG and Deutsche Telekom AG ("Deutsche Telekom"), the fulfilment of which requires T-Mobile to issue, for no additional consideration, an aggregate of 48,751,557 shares of common stock of T-Mobile (the "Contingent Consideration") to SBG or its designated affiliate.

The final accounting was surprising: Including the Shares, the investment in Sprint reached an equity IRR (internal rate of return) of 25.5% and a MOIC (multiple of invested capital) of 8.1 times as of December 22, 2023. The merger with T-Mobile, combined with T-Mobile's subsequent stock appreciation, transformed a operational disaster into a financial success. Son had failed to build America's leading carrier but succeeded in capturing value from someone else doing it.

The Sprint saga encapsulated everything about Son's investment style: the audacious vision, the massive capital deployment, the operational struggles, and the financial engineering that somehow produced a profit. It also revealed the limits of the SoftBank playbook. In mature markets with established competitors and complex regulations, money alone couldn't buy market leadership. You needed operational excellence, regulatory savvy, and most importantly, timing.

Sprint also marked a turning point in Son's thinking. The seven-year struggle to fix an operating business while simultaneously trying to merge it convinced him that SoftBank's future lay not in operating companies but in pure investment. The Vision Fund, announced while Sprint was still bleeding cash, represented Son's conclusion from the Sprint experience: why struggle to run one company when you could own pieces of hundreds?

Looking back, Sprint was both Son's greatest failure and most important education. It failed as a business turnaround, failed as a merger strategy, and failed to create the global telecom empire Son envisioned. But it succeeded in teaching Son that in technology investing, it's better to bet on entrepreneurs than to become one. The next chapter of SoftBank would be built on that insight.

↑ Back to Top

VII. ARM Acquisition: The $32B Semiconductor Play (2016)

The private jet from Tokyo touched down at London City Airport on a gray July morning in 2016. Masayoshi Son had been in the air for twelve hours, but he wasn't tired. He was electric with excitement. In two weeks, he had conceived, negotiated, and was about to announce the largest technology acquisition in European history. ARM Holdings, the crown jewel of British technology, would be his for £24.3 billion ($32 billion) in cash. The entire negotiation, from first contact to signed deal, had taken just fourteen days. Even by Son's standards, this was audacious.

Japan's Softbank has agreed to acquire ARM Holdings, the giant U.K. semiconductor firm that supplies part of the chip design used in Apple iPhones, in a deal worth more than $32 billion. Japanese conglomerate SoftBank Group made an agreed offer for ARM on 18 July 2016, subject to approval by ARM's shareholders, valuing the company at £23.4 billion (US$32 billion). The transaction was completed on 5 September 2016.

The acquisition made no sense to conventional investors. ARM didn't manufacture chips—it designed them and licensed the designs. Its 2015 revenue was just $1.5 billion, meaning Son was paying over 20 times revenue. ARM's growth rate was solid but unspectacular. And most puzzling: this was a complete departure from SoftBank's internet and telecom focus. When the news broke, SoftBank's stock immediately dropped 10% as investors struggled to understand why a Japanese telecom-internet conglomerate was buying a British semiconductor design company.

But Son saw something others missed. Every smartphone in the world contained ARM-designed processors—over 95% market share. ARM's architecture powered everything from Apple's A-series chips to Qualcomm's Snapdragon processors. ARM, the largest London-listed tech company by market value, is a major presence in mobile processing, with its processor and graphics technology used by Samsung, Huawei, and Apple in their in-house microchips. Components based on technology licensed by ARM are found in the vast majority of the world's smartphones.

The real opportunity, however, wasn't smartphones—it was what came next. Son told reporters that the acquisition was his "big bet for the future," explaining that ARM was well positioned to capitalize as more and more devices -- household appliances, cars and censors -- are connected to the Internet. IoT connected devices will number 38.5 billion in 2020, up from 13.4 billion in 2015, a rise of over 285 percent, according to Juniper Research, highlighting the interest in this sector from major technology giants. Masayoshi Son described the IoT as a "paradigm shift".

The timing of the acquisition was no accident. Brexit had just happened, crushing the British pound by 15%. Son was essentially getting a massive discount courtesy of currency movements. But he insisted this wasn't opportunistic: "This is not opportunistic about the currency," he joked. "I have wanted to do this but was waiting for the cash to come in. I'm not investing in a distressed asset," he continued. "I'm investing in a paradigm shift…. that's my passion, and that's my view."

The negotiation itself was vintage Son. Simon Segars, ARM's CEO, later revealed that ARM wasn't even looking to sell. Son called him directly, flew to Cambridge, and made an offer so compelling that the board couldn't refuse. As part of its bid to bolster its presence in the growing internet of things (IoT) sector, Softbank will pay £17 per share (about $22.50 a share) for ARM, a 43 percent premium on Friday's closing price, in an all-cash deal. The deal works out to 1,700 pence per ARM Share, a premium of 43.0 percent on the closing price of 1,189 pence per ARM Share on July 15, 2016.

To sweeten the deal and address political concerns about a British crown jewel being sold to Japan, Son made unprecedented commitments. It lets ARM to stay headquartered in Cambridge and promises to double the company's staff from its current complement of 4,000 employees over the next five years. This wasn't just PR—Son genuinely believed ARM needed more engineers to capture the IoT opportunity.

The strategic logic was compelling once you understood Son's worldview. He believed that computing was evolving through three phases: PC era (dominated by Intel), mobile era (dominated by ARM), and IoT era (up for grabs). If Softbank was in the thick of it at the "beginning of the PC internet," in Masayoshi's words (he invested in via Yahoo when it only had 16 employees), and then doubled down on mobile (by way of Softbank's investments in various mobile companies and acquisition of Sprint) then — he believes — IoT is the natural progression of that.

Son's vision for IoT went far beyond smart thermostats and fitness trackers. He imagined a world where everything had intelligence: "I would say automobile is becoming smarter and smarter so when automobile becomes so smart it is required to have more and more chips integrated inside the car, especially when it becomes a driverless car. Automotive itself will become a super computer which consists of a bunch of multiple chips so ARM will be going into that market very aggressively," Masayoshi Son said.

ARM's business model was particularly attractive to Son. Unlike Intel, which manufactured its own chips, ARM licensed designs and collected royalties—a capital-light model with 40% operating margins. Every new connected device would need processors, and most would use ARM designs. It was like owning a toll booth on the information superhighway, collecting a few cents from every smart device sold globally.

The company also came with hidden assets. Late last year, ARM announced a deal with IBM to collaborate via IBM's IoT Foundation to use ARM devices based on its "mbed" chips to collect and analyze data from intelligent devices like industrial appliances, weather sensors and monitoring devices. ARM's IoT strategy focuses on developing and scaling its "mbed" technology, which includes a "full-stack" operating system tailored to its Cortex-M 32-bit microcontrollers and a "device server" that handles connections from IoT devices.

The financing structure revealed SoftBank's financial position post-Sprint. There were other financial factors in play as well: in the weeks leading up to the deal, Softbank sold a chunk of its Alibaba stake and its stake in Supercell, and today it announced a large loan for some $9 billion (¥1 trillion). Son was essentially recycling capital from mature investments into what he saw as the next platform shift.

But the ARM acquisition also exposed the tension at the heart of SoftBank. The deal has taken some heat from a number of financial analysts, many of whom believe the transaction will burden the company with too much debt – about $10 billion dollars more when the acquisition completes. A related worry is there will be less funds available to invest in Sprint, the US wireless carrier SoftBank has a controlling stake in. Given SoftBank's tumbling stock price since the news was announced, many investors seem to agree the deal is risky. In fact, some of the company's shareholders are balking at the move.

The board tensions were real. Nikesh Arora, Son's heir apparent and the architect of SoftBank's investment strategy, had resigned just weeks before the ARM deal. There have been major changes to Softbank's leadership recently. Superstar executive Nikesh Arora, who was being groomed as Son's successor, resigned as company president in June. His departure signaled that Son, then 58, wasn't ready to hand over control and was about to embark on his most ambitious phase yet.

Today, Softbank announced that it has completed its acquisition of ARM Holdings, the semiconductor firm that it said in July it would acquire for £24 billion in cash (around $32 billion in today's currency, $31 billion at the time of the deal), in order to make a big jump into IoT. The acquisition closed on September 5, 2016, transforming SoftBank from a telecom-internet company into a company that owned the fundamental architecture of mobile computing.

The immediate aftermath was anticlimactic. ARM continued operating as before, designing chips and collecting royalties. Revenue grew steadily but not spectacularly. The IoT explosion Son predicted was slower to materialize than expected. By 2019, with the Vision Fund struggling and SoftBank needing cash, Son began exploring options.

The attempted sale to Nvidia in 2020 would have been the ultimate validation—a $40 billion acquisition becoming a $66 billion sale in just four years. American technology company Nvidia announced plans on 13 September 2020 to acquire ARM from SoftBank, pending regulatory approval, for a value of US$40 billion in stock and cash. However, the European Commission, the UK Competition and Markets Authority and the US Federal Trade Commission raised completion concerns focusing on Arm's role within Nvidia, while the UK government also raised concerns about national security. The merger attempt was eventually cancelled in February 2022 due to the aforementioned regulatory pressure and hurdles.

The regulatory rejection of the Nvidia deal forced Son to pivot to Plan B: an IPO. Arm went public on 14 September 2023 raising $4.87 billion at a $54.5 billion valuation, with SoftBank continuing to own roughly 90% of the company following the offering. The valuation vindicated Son's thesis—ARM was worth 70% more than he paid, even after the tech wreck of 2022.

But the real test of the ARM acquisition isn't financial returns—it's strategic positioning. The acquisition of ARM Holdings in 2016 for $32 billion has significantly impacted SoftBank. Since the acquisition, SoftBank's share price has increased by approximately 78% year-to-date, largely due to ARM's successful IPO. SoftBank retains about 90% ownership of ARM post-IPO. ARM's shares have surged nearly 124% this year as of July 3, indicating strong market performance and investor confidence. The acquisition has strengthened SoftBank's position in the technology sector, particularly in the IoT market, and has allowed ARM to invest more in research and development, leading to innovations in chip technology and expansion into new markets.

The ARM story also illustrates Son's evolution as an investor. Unlike the operational struggles with Sprint or the venture capital chaos of WeWork, ARM was a strategic acquisition of a profitable, well-run company in a dominant market position. Son didn't try to transform ARM or accelerate its growth artificially. He simply owned it, funded its R&D, and waited for the market to recognize its value.

"We have long admired ARM as a world renowned and highly respected technology company that is by some distance the market-leader in its field," said Masayoshi Son, chairman and CEO of SoftBank. "ARM will be an excellent strategic fit within the SoftBank group as we invest to capture the very significant opportunities provided by the Internet of Things. This is one of the most important acquisitions we have ever made, and I expect ARM to be a key pillar of SoftBank's growth strategy going forward."

In retrospect, the ARM acquisition was Son's most prescient bet. Not because of IoT, which has been slower to develop than predicted, but because of AI. Every AI inference chip, every edge computing device, every neural processing unit needs an architecture—and increasingly, that architecture is ARM. The acquisition Son made to capture the IoT revolution ended up positioning SoftBank at the heart of the AI revolution. Sometimes in investing, being right for the wrong reasons is the best kind of right.

↑ Back to Top

VIII. Vision Fund Era: Redefining Venture Capital (2017-2019)

The Ritz-Carlton in Riyadh was buzzing with energy in October 2016. Tech executives, Saudi princes, and investment bankers mingled in the opulent lobby, but the real action was happening in a private suite upstairs. Masayoshi Son sat across from Crown Prince Mohammed bin Salman, sketching on a napkin his vision for the future of technology. "Three hundred years," Son said, his eyes gleaming. "We're building a three-hundred-year vision for the information revolution." MBS leaned forward, intrigued. In forty-five minutes, they shook hands on what would become the largest technology investment fund in human history.

​​SoftBank Group Corp. ("SBG") today announced that it will form SoftBank Vision Fund (tentative name) (the "Fund"). The Fund intends to make investments in the technology sector globally. The Fund will be managed in the United Kingdom by a subsidiary of SBG and will deploy capital from SBG and investment partners. The Fund will aim to be one of the world's largest of its kind. The overall potential size of the Fund can go up to USD 100 billion.

SBG expects to invest at least USD 25 billion over the next 5 years. SBG has concluded a non-binding memorandum of understanding ("MOU") on October 12, 2016 with the Public Investment Fund of the Kingdom of Saudi Arabia ("PIF"), under which PIF will consider investing in the Fund and becoming the lead investment partner, with the potential investment size of up to USD 45 billion over the next five years.

The Vision Fund wasn't just big—it was incomprehensibly big. The SoftBank Vision Fund is a venture capital fund founded in 2017. It is managed by SoftBank Investment Advisers, a subsidiary of the SoftBank Group. With over $100 billion in capital, it is the world's largest technology-focused investment fund. To put this in perspective, the entire U.S. venture capital industry invested about $70 billion annually. Son had raised more money for a single fund than the entire American VC ecosystem deployed in a typical year.

The fund's creation story was as audacious as its size. The Softbank Vision Fund was created in May 2017 by the SoftBank Group and the Public Investment Fund (PIF). $100 billion was raised with PIF contributing $45 billion, SoftBank contributing $28 billion, the Mubadala Investment Company contributing $15 billion and the rest from other investors including Apple. The investor list read like a who's who of global capital: Saudi Arabia's sovereign wealth fund, Abu Dhabi's Mubadala, Apple, Foxconn, Qualcomm, Sharp.

But the most remarkable aspect wasn't the size or the investors—it was the speed. Riyadh, 20 May 2017 - The SoftBank Vision Fund (the "Vision Fund") today announced that in addition to SoftBank Group Corp. ("SBG") and the Public Investment Fund of the Kingdom of Saudi Arabia ("PIF"), investors in the Vision Fund include the Mubadala Investment Company of the United Arab Emirates ("Mubadala"), Apple Inc ("Apple"), Foxconn Technology Group ("Foxconn"), Qualcomm Incorporated ("Qualcomm") and Sharp Corporation ("Sharp"). Following the first major close, the Vision Fund has over $93 billion of committed capital. The Fund is targeting a total of $100 billion of committed capital, with a final close within six months.

Son's pitch to investors was simple and grandiose: Through Softbank Vision Fund, Masayoshi Son explained his intent to invest in all companies developing technologies in line with the global artificial intelligence trends, including various sectors such as finance or transportation. He wasn't thinking about individual companies or sectors—he was thinking about owning the entire stack of the AI revolution.

The philosophy behind the Vision Fund was radically different from traditional venture capital. Where VCs wrote $10 million checks and hoped for 10x returns, Son wrote $1 billion checks and aimed to create monopolies. His theory: in winner-take-all markets, the company with the most capital wins. Give Uber enough money to subsidize rides indefinitely, and it would destroy all competition. Give WeWork enough capital to lease every office building in Manhattan, and it would become the only office provider that mattered.

Masayoshi Son, Chairman & CEO of SoftBank Group Corp. said: "Technology has the potential to address the biggest challenges and risks facing humanity today. The businesses working to solve these problems will require patient long-term capital and visionary strategic investment partners with the resources to nurture their success. SoftBank has long made bold investments in transformative technologies and supported disruptive entrepreneurs. The SoftBank Vision Fund is consistent with this strategy and will help build and grow businesses creating the foundational platforms of the next stage of the Information Revolution."

The operational structure of the Vision Fund was as unconventional as its size. The new fund, dubbed the SoftBank Vision Fund, will be based in London and seeded with $25 billion from SoftBank and up to $45 billion from Saudi Arabia's sovereign wealth fund over the next five years, according to a statement from Masayoshi Son's Japanese telecoms group. Based in London, run by ex-Deutsche Bank trader Rajeev Misra, the fund operated more like a sovereign wealth fund than a traditional VC.

The investment pace was frenetic. Within months of closing, the Vision Fund was deploying capital at a rate never before seen in venture capital. On 25 August 2017, SoftBank finalized a $4.4 billion investment in WeWork. On 24 October 2017, Son announced the group would collaborate with Saudi Arabia to develop Neom, the new high-tech business and industrial city of the Saudi Kingdom. On 14 November 2017, Softbank agreed to invest $10 billion into Uber. On 29 December 2017, it was reported that a SoftBank-led consortium had invested $9 billion into Uber. The deal, to close in January 2018, would leave SoftBank as Uber's biggest shareholder, with a 15 percent stake. The deal was secured after Uber shareholders voted to "sell their shares to the Japanese conglomerate at a discounted price."

The Vision Fund's investment strategy was to identify category leaders and give them so much capital they couldn't lose. Uber got $9 billion. WeWork got $4.4 billion initially, then billions more. DoorDash, Slack, ByteDance—if you were the leader in a large market, the Vision Fund wanted to give you money. The fund didn't just participate in rounds; it created its own rounds, often at valuations 30-50% higher than what other investors were willing to pay.

This approach fundamentally changed Silicon Valley's dynamics. Suddenly, startups that would have raised $50 million were raising $500 million. Companies that would have gone public at $1 billion valuations were staying private at $10 billion valuations. The Vision Fund wasn't just participating in the venture capital market—it was remaking it in its own image.

The fund's portfolio construction followed Son's worldview about the AI revolution. He invested in the infrastructure layer (ARM, Nvidia), the application layer (Uber, DoorDash, WeWork), and the AI layer itself (various AI startups). The thesis was that these companies would eventually integrate, creating a seamless AI-powered ecosystem that the Vision Fund would control.

But there were early warning signs. The Saudi connection created complications. When SoftBank invested $5 billion in Chinese ride-hailing company Didi Chuxing last year, it created a separate investment vehicle, called the Delta Fund, to complete the transaction. The purpose was to carve out an investment away from the Saudi PIF, which had already put $3.5 billion in Didi rival Uber and wanted to avoid funding a competitor, according to two people familiar with the matter, who asked not to be named because the strategy is confidential.

The pace of investment was unsustainable. The fund had invested $23.5 billion of about $92 billion in committed capital as of June 30. That number, which does not include more than $20 billion in ride-sharing investments, will jump to about $40 billion when SoftBank next reports earnings on Nov. 5, according to people with knowledge of the company. There's about $35 billion to $40 billion left to invest out of Vision Fund 1.

More fundamentally, the Vision Fund's strategy of creating monopolies through capital injection was running into reality. Uber was burning billions but still losing to Lyft in key markets. WeWork was leasing buildings at a loss with no path to profitability. The Chinese government was cracking down on tech monopolies, threatening investments in Didi and ByteDance.

The fund's governance structure was also problematic. PIF and Mubadala receive a 7 percent dividend on invested capital plus a percentage of any returns, the report explains. This preferred return meant that even if the fund broke even, SoftBank would lose money after paying out the preferreds. The incentive structure encouraged increasingly risky bets to generate returns above the hurdle rate.

By 2019, cracks were showing. Portfolio companies were struggling to justify their valuations. The IPO window was closing as public markets rejected money-losing unicorns. In January 2020, multiple Softbank-funded startups started cutting their staff, such as at Getaround, Oyo, Rappi, Katerra and Zume.

The Vision Fund represented the logical extreme of Son's investment philosophy: if capital is the key to winning technology markets, then infinite capital should guarantee victory. But the fund's early years would prove that even $100 billion couldn't overcome bad business models, poor governance, and market reality. The Vision Fund had redefined venture capital—the question was whether that redefinition was sustainable.

Shortly after his election, President Donald Trump, after meeting with Son at Trump Tower in New York in December, said the CEO had committed to investing $50 billion in the United States with the goal of creating 50,000 jobs. Although Trump claimed Son said that intended investment wouldn't have happened without Trump election, Dow Jones reported that the money would come from the Vision Fund.

Looking back, the Vision Fund's 2017-2019 era was peak Son: audacious in vision, unprecedented in scale, transformative in impact, and heading toward a reckoning. The fund had achieved its goal of changing how technology companies were funded. Whether that change was positive remained to be seen. As one Silicon Valley veteran put it: "Masa didn't just bring capital to the party—he brought a flamethrower to a candlelit dinner."

↑ Back to Top

IX. WeWork Debacle: The $47B to $2.9B Implosion (2017-2020)

The first time Masayoshi Son met Adam Neumann, the meeting lasted twelve minutes. It was 2016 in India, and Neumann had cornered Son at a conference, launching into his pitch with characteristic intensity. "We're not a real estate company," Neumann insisted, his eyes blazing with conviction. "We're a physical social network. We're consciousness. We're energy. We're about elevating the world's consciousness." Son, who had heard thousands of pitches, was transfixed. Later, he would tell others that Neumann reminded him of a young Jack Ma—charismatic, crazy, and possibly brilliant. That twelve-minute meeting would eventually cost SoftBank $13 billion.

SoftBank has put money in WeWork several times, including a $4.4 billion investment in 2017 at a valuation of about $20 billion. Commercial office space start-up WeWork announced it has $4.4 billion from SoftBank and its related Softbank Vision Fund, which includes other investors. The deal includes $1.4 billion earmarked to help WeWork expand throughout China, Japan and Southeast Asia, as well as $3 billion in growth funding into WeWork's parent company. The investment represented the Vision Fund's first major deployment—a signal of the kind of companies Son wanted to back.

The relationship between Son and Neumann was unlike typical investor-founder dynamics. In 2016, Son met Neumann in India. Son and Neumann shared a love of taking big risks. SoftBank first committed $3.1 billion in new funding to WeWork in 2017. "Neumann has told others that Son appreciated how he was crazy–but thought that he needed to be crazier," according to the Journal. This wasn't just venture capital speak—Son genuinely believed that only the truly insane could build trillion-dollar companies.

Personally and through his Vision Fund, Son invested about $9 billion in WeWork — most recently at a valuation of $47 billion in January, according to the Wall Street Journal. In January 2019, WeWork raised an additional $2 billion from SoftBank at a $47 billion valuation. The valuation made no sense by any traditional metric. WeWork competitor IWG has a market capitalization of $3.6 billion on trailing 12-month revenue of $2.7 billion, a 1.3x price-to-revenue multiple. WeWork's comparable revenue as of June 30, 2019, was $2.6 billion. A valuation of $47 billion marked SoftBank at an 18x price-to-revenue multiple.

But Son wasn't investing based on traditional metrics. He saw WeWork as the physical manifestation of the sharing economy—Uber for office space. Neumann's pitch for money — he was bringing to life "a new way of work to a changing world" which included a mobile app that would "facilitate a physical social network".– was effective. The vision was intoxicating: WeWork would become the world's first physical social network, connecting millions of entrepreneurs and reshaping how humanity worked.

The business model, however, was fundamentally flawed. WeWork signed long-term leases (typically 15 years) and rented space short-term (month-to-month or annual). This arbitrage worked in good times but created massive risk in downturns. It also disclosed $47 billion of future lease obligations and only $4 billion of future lease commitments. The company was essentially a real estate company with a tech valuation—a dangerous combination that required suspension of disbelief.

Neumann's behavior became increasingly erratic as the valuations soared. The Wall Street Journal reported that he had taken $700 million out of WeWork before the IPO, among other details, and "undermined his position" at the company. Neumann also directed We Holdings LLC (a company managed by Neumann and McKelvey) to unwind the transaction of $5.9 million in stock that the company paid in exchange for the "We" trademarks. He smoked marijuana on private jets, made employees do shots of tequila at mandatory company events, and talked about becoming president of the world and living forever.

Within SoftBank, there were dissenting voices. At one point, Son's optimism about WeWork was countered by dissenting voices, such as Nikesh Arora and Alok Sama, according to people familiar with the matter. Key internal dissenters on WeWork, including Nikesh Arora and Alok Sama, have departed SoftBank in recent years. Arora and Sama looked at WeWork several years ago and advised not to invest at an $8 billion valuation, sources say. But Son overruled them all. He saw something in Neumann that transcended spreadsheets.

The unraveling began with the S-1 filing in August 2019. On August 14, 2019, the company filed Form S-1 in preparation for an initial public offering. The filing revealed significant losses, expensive lease agreements, and a complex relationship with founder Adam Neumann. The company was then "besieged with criticism over its governance, business model, and ability to turn a profit." WeWork's public prospectus filing in August showed $900 million in losses over the first six months of 2019, revealing major cracks in the company's business model that eventually sank the planned IPO before it launched.

The public market's reaction was brutal. Investors saw what SoftBank had willfully ignored: a company losing $2 billion annually with no path to profitability, a CEO who had extracted hundreds of millions while employees held worthless options, and governance so bad it became a Harvard Business School case study in what not to do. On September 17, 2019, amid growing investor concerns over its corporate governance, valuation, and outlook for the business, WeWork formally withdrew its S-1 filing and announced the postponing of its IPO. At that time, the reported public valuation of the company was around $10 billion, a reduction from the $47 billion valuation it achieved in January and less than the $12.8 billion it had raised since 2010.

The speed of the collapse was breathtaking. That's still far more than the likely public market valuation of WeWork, which had fallen as low as $10 billion, Reuters reported two weeks ago. In less than a month, WeWork went from a $47 billion valuation to near bankruptcy. By September 23, 2019, SoftBank wanted Neumann removed as chief executive. Following mounting pressure from investors based on disclosures made in a public offering filing, Neumann was asked to step down as CEO of WeWork and gave up majority voting control as of September 26, 2019. On September 24, 2019, he resigned and Artie Minson and Sebastian Gunningham were named as successors.

Son faced a terrible choice: let WeWork fail and lose everything, or throw good money after bad. He chose the latter. Still, in October, SoftBank said it would bail out WeWork by investing another $3 billion that would give the Japanese company control of the start-up. The deal valued WeWork at roughly $8 billion, or less than one-fifth the company's valuation just eight months earlier.

The bailout terms were both generous to Neumann and insulting to everyone else. In October 2019, The Wall Street Journal reported that Neumann would receive close to $1.7 billion from stakeholder SoftBank for stepping down from WeWork's board and severing most of his ties to the company. To recover credibility, investors needed Adam out, but his exit was costly. He received $185 million for a non-compete agreement, a $106 million settlement payment, and an extended $432 million loan. When WeWork went public in 2021, SoftBank bought $480 million of his shares, fulfilling a contractual obligation.

The aftermath was devastating for SoftBank's reputation. At the same time, SoftBank was also trying to raise billions more for a second Vision Fund — and printing a headline valuation of $47 billion for WeWork would signal the first fund was firing on all cylinders. Son claimed his Vision Fund limited partners had already earned a 45% return on their money on Vision Fund 1 earlier this year — but that included paper gains, such as WeWork and Uber, another large SoftBank investment, which has lost value since going public.

Son's explanation was revealing: Masayoshi Son's genuine optimism about WeWork, and his perception that others would agree, led SoftBank to be comfortable with a $47 billion valuation in January, sources say. Those close to Son say his faith in WeWork isn't based on misdirection as much as it's founded in his belief that the company will turn into a massive financial success with time and others would view it in a similar light. He genuinely believed WeWork would revolutionize work. The fact that public markets disagreed was, to him, their failure of imagination, not his.

In March 2020, SoftBank announced that SoftBank gave WeWork a valuation of $2.9 billion as of March 31 based on a discounted cash flow method, down from $7.3 billion as of Dec. 31, following its failed IPO. The $47 billion valuation had become $2.9 billion in just over a year—a 94% decline that represented one of the most spectacular destructions of value in corporate history.

WeWork's botched plan to go public is the latest example of financial realities clashing with Masayoshi Son's 300-year vision and the long-term investment thesis of his $100 billion Vision Fund. The WeWork debacle revealed the fatal flaw in the Vision Fund model: when you have $100 billion to deploy, you stop seeing red flags—you only see opportunities. The due diligence that would have caught Neumann's self-dealing, the analysis that would have questioned the business model, the governance requirements that would have prevented the chaos—all were sacrificed in the rush to deploy capital.

SoftBank and its Vision Fund are WeWork's biggest investors, with a stake of around 29% in We Co., the company's official name. That support is a key reason why this real-estate rental company fetched a $47 billion valuation. At one point SoftBank founder Masayoshi Son described WeWork as the next Alibaba Group Holding Ltd. in reference to the fortune he made from investing in Jack Ma's Chinese e-commerce giant.

Looking back, WeWork represented peak Vision Fund hubris. It was the moment Son's reality distortion field collided with actual reality—and reality won. The company that was supposed to elevate the world's consciousness instead became a cautionary tale about what happens when too much capital meets too little discipline. As one SoftBank executive later admitted privately: "We didn't invest in WeWork. We invested in Adam Neumann's personality. That was our mistake."

↑ Back to Top

X. The COVID Crisis & Recovery (2020-2021)

The boardroom at SoftBank's Tokyo headquarters was silent on March 13, 2020. Masayoshi Son stood before a single slide showing SoftBank's stock price—down 50% in three weeks. The Vision Fund portfolio was hemorrhaging value. WeWork was on life support. Uber and DoorDash were collapsing as lockdowns killed demand. Elliott Management, the activist fund that had taken a $2.5 billion stake, was demanding blood. Son, who had survived the dot-com crash and the financial crisis, looked at his executives and said simply: "We go to war." What followed was the most dramatic financial engineering in SoftBank's history.

In March 2020, SoftBank announced that it was launching an emergency ¥4.5tn ($41bn) asset sale to fund a share buyback and debt reduction. The effort was initiated by Son in order to stem a collapse in the company's share price due to the pandemic. TOKYO (Reuters) - SoftBank Group Corp <9984.T> plans to raise as much as $41 billion to buy back shares and reduce debt in an unprecedented move to restore investor confidence as a financial market rout pummels its shares and its portfolio companies.

The scale of the crisis was unprecedented. The Japanese company said it expected an operating loss of 1.35 trillion yen ($12.5 billion) in its fiscal year through March 31, 2020. That compares with an operating profit of more than 2 trillion yen the previous year. The Vision Fund is expected to report losses amounting to 1.8 trillion yen ($16.7 billion) due to "the deteriorating market environment," the company said in a statement.

The asset sale program was Son at his most decisive. Its plans come as it contends with a growing financial squeeze on the company and its $100 billion Vision Fund, which has recorded two consecutive quarters of losses after its tech bets fell short, compounded by the coronavirus pandemic's impact on the global economy. The buyback tops the $20 billion of purchases sought by activist investor Elliott Management, which has put pressure on SoftBank to improve shareholder returns, and will retire 45% of the group's shares.

The execution was ruthless. Within weeks, SoftBank was selling everything that wasn't nailed down. Alibaba shares, T-Mobile stakes, pieces of its telecom business—nothing was sacred. The message to the market was clear: SoftBank had the liquidity to survive anything. After the programme was unveiled, Softbank share price rose almost 19%. The program included a plan to repurchase ¥2tn of its shares in addition to the ¥500bn buyback it promised 10 days prior. Combined, SoftBank would be repurchasing 45% of its stock.

But the real drama was happening inside the Vision Fund. In January 2020, multiple Softbank-funded startups started cutting their staff, such as at Getaround, Oyo, Rappi, Katerra and Zume. Portfolio companies were burning through cash with no access to new capital. The pandemic had exposed the fundamental weakness of the Vision Fund model: when you give companies unlimited capital to grow at any cost, they become addicted to that capital. Take it away, and they collapse.

In February 2020, Elliott Management, an activist hedge fund, bought a $2.5 billion stake in Softbank and pushed for restructuration and more transparency, especially regarding its Vision Fund. Elliott's demands were specific: sell assets, buy back stock, improve governance, and most importantly, stop making new Vision Fund investments. It was essentially demanding that Son abandon his life's work.

The portfolio carnage was spectacular. Masayoshi Son's company expects its $100 billion Vision Fund— which includes a massive investment in WeWork — to record a $16.6 billion loss for the fiscal year, according to the Wall Street Journal. For Softbank, a big share of thoses woes stem from WeWork's failed IPO late last year, and the co-working firm's skyrocketing debt. But overall, SoftBank blamed Vision Fund's enormous losses on a "deteriorating market environment," the Journal reported.

The irony was that COVID should have been the Vision Fund's moment. The pandemic accelerated digital transformation by a decade. E-commerce, food delivery, remote work, digital payments—every thesis the Vision Fund had bet on was suddenly mainstream. But the portfolio companies were too damaged to capitalize. They had been built for growth, not survival.

Then came the unexpected reversal. By summer 2020, central banks had flooded markets with liquidity. Tech stocks began soaring. The SoftBank Vision Fund, which reported an $18 billion loss last year, recorded an investment gain of $2.8 billion for this quarter. SoftBank Group posted a $12 billion quarterly profit on Tuesday after reporting historic losses of $13 billion for its last fiscal year. The recovery is largely down to rising valuations of SoftBank Vision Fund bets like Uber and Slack, which saw their share prices rally during the April to June quarter as U.S. tech stocks soared during the coronavirus pandemic.

The rally accelerated through 2020 and into 2021. DoorDash went public at $102 per share in December 2020, giving SoftBank a massive gain on its investment. Coupang's IPO in March 2021 was even better—the Korean e-commerce company's debut created billions in paper profits. The climax came in March 2021. In March 2021, SoftBank made a record $36.99 billion profit from its Vision Fund unit and investment gains via the public market debut of Coupang. SoftBank Group's net profit was $45.88 billion (¥4.99 trillion). It was the largest recorded annual profit by a Japanese company in history.

The numbers were staggering. Group net profit was 4.99 trillion yen ($45.88 billion) in the year ended March, topping the $42.5 billion made by Warren Buffett's Berkshire Hathaway in its last business year. The annual profit is the largest posted by a Japanese company to date. The group's $45.88 billion annual profit propelled it to third place on the list of the world's most profitable companies, after Apple and Saudi Aramco.

The Coupang IPO was the crown jewel. The South Korean ecommerce juggernaut, debuts on Wall Street in a deal that could value the company at more than $51 billion on a fully diluted basis. Assuming it goes out at that level, the Coupang IPO is poised to be the largest for a foreign company in the US since Alibaba, according to PitchBook data. Moreover, it would make the Vision Fund's 33% stake in Coupang worth roughly $16 billion, far larger than its holdings in Uber or DoorDash.

Son's victory lap was characteristically grandiose. In May 2021, SoftBank announced the total fair value of both funds as of 31 March 2021 was $154 billion and the Vision Funds made a record profit of $36.99 billion due to its successful investment in Coupang. After announcing the success, SoftBank raised the size of Vision Fund 2 to $30 billion and stated it plans to continue self funding the second fund although it might consider trying again to secure funding from external investors.

The comeback narrative was irresistible. A year earlier, SoftBank had been on the brink of collapse, forced to sell assets at fire-sale prices. Now it was posting the largest profit in Japanese corporate history. Son, who had been written off as a reckless gambler, was vindicated. The Vision Fund, declared dead after WeWork, was printing money.

But there were warning signs even in triumph. The profits were largely paper gains on public market valuations that could evaporate as quickly as they appeared. Many portfolio companies were still losing money operationally. The IPO window that had been so generous could slam shut at any moment. And most ominously, Son was already talking about his next big bet: artificial intelligence.

The COVID crisis had taught Son several crucial lessons. First, massive liquidity injections by central banks could bail out even the worst investments—if you could survive long enough. Second, the Vision Fund model of growth at any cost was fundamentally flawed in a world where capital could suddenly disappear. Third, and most importantly, SoftBank needed to be less dependent on external capital.

This last lesson would drive Son's next evolution. The Vision Fund 2, announced with much fanfare as a $108 billion fund, ended up with just $30 billion—all from SoftBank itself. The Saudis and Emiratis who had funded the first Vision Fund were nowhere to be seen. Son was going it alone, which meant smaller bets but more control.

The recovery also masked deeper structural issues. While the headline profit number was spectacular, the underlying business performance of many portfolio companies remained weak. DoorDash was still losing money on every delivery. Uber's path to profitability remained unclear. The "harvesting period," as Son calls it, began with a bumper crop of 11 exits since last year. But what happened when there was nothing left to harvest?

In retrospect, the COVID crisis and recovery represented both SoftBank's greatest trial and its most Pyrrhic victory. The company survived and even thrived, but at the cost of its original vision. The Vision Fund was no longer about transforming industries or building the future—it was about financial engineering, riding market waves, and hoping valuations held up long enough to exit.

"This programme will be the largest share buyback and will result in the largest increase in cash balance in the history of SBG [SoftBank Group], reflecting the firm and unwavering confidence we have in our business," Son had said in March 2020. By March 2021, that confidence seemed justified. But as every trader knows, the market can remain irrational longer than you can remain solvent—and it can also become rational faster than you can exit.

↑ Back to Top

XI. Vision Fund Reckoning & Strategic Pivot (2022–Present)

The reckoning came faster than anyone expected. In May 2022, SoftBank's earnings call felt more like a funeral than a quarterly update. Masayoshi Son, usually ebullient and armed with slides about the future, stood before a single chart showing a sea of red. "We are in a defensive mode," he said quietly. The numbers were catastrophic: In 2022, SoftBank Vision Fund posted a record 3.5 trillion yen loss ($27.4 billion) for its financial year ended on 31 March 2022 as the valuation of its stock portfolio plummeted.

The scale of the destruction was unprecedented. In May 2023, the SoftBank Group disclosed that its Vision Fund lost a record $32 billion in the fiscal year ending in March 2023. The fund that had defined an era of venture capital excess was now its biggest casualty. Eric J. Savitz, associate editor for technology at Barron's, characterized the SoftBank Vision Fund as a failed experiment.

The portfolio carnage was universal. Every assumption that had driven Vision Fund investments—that growth was more important than profits, that winner-take-all dynamics justified any valuation, that technology would eat every industry—was being rejected by public markets. Interest rates were rising, ending the era of free money. Investors wanted profits, not promises.

By early 2023, having lost its exuberance due to serious profitability issues and facing declining return on investment prospects as well as record losses, the once world's biggest investor in startups had invested just $300 million into startups in the last October–December quarter, down more than 90% from the previous year. The Vision Fund had gone from writing billion-dollar checks to effectively shutting down new investments.

The human toll was significant. The fund's losing investments were massive and both Son, who had claimed he had a special ability to see the future of disruptive entrepreneurship, and SoftBank, received much criticism from shareholders, peers and the business and finance media in general due to the magnitude of the failure. Costly failures like those of Katerra, Wirecard and Zymergen were just a few examples of SoftBank's reckless investing behaviour and the investing holding company's incompetence, neglect and failure to show due diligence.

The leadership exodus accelerated. In July 2022, CEO Rajeev Misra announced he would be stepping back from some of his main roles including the management of SoftBank Vision Fund 2. Several other executives stepped down from their roles in the same year. In November 2024, Misra formally stepped down from his role in managing the Vision Fund and will be succeeded by Alex Clavel. The fund that had once employed 500 people was down to a skeleton crew.

Son's response was a complete strategic pivot. Instead of trying to fix the Vision Fund model, he abandoned it. The grand vision of launching a new $100 billion fund every few years was dead. Vision Fund 2, which was supposed to be even larger than the first, raised just $30 billion—all from SoftBank itself. There would be no Vision Fund 3.

The new strategy was radically different: focus on a handful of large, strategic investments rather than spray-and-pray across hundreds of startups. The biggest bet was on artificial intelligence, particularly through ownership of ARM and new investments in AI infrastructure. Son was essentially admitting that the Vision Fund experiment had failed and returning to SoftBank's roots as a strategic investor.

The ARM IPO in September 2023 symbolized this shift. Arm went public on 14 September 2023 raising $4.87 billion at a $54.5 billion valuation, with SoftBank continuing to own 90.6% of the company following the offering. Rather than using ARM as currency for more Vision Fund investments, Son held onto it as a core strategic asset. The message was clear: SoftBank was done being a venture capitalist.

The asset sales accelerated. Since the end of 2021, the world's biggest startup fund has seen its U.S.-listed portfolio shrink by almost $29 billion, as it sold down stakes in companies such as Coupang Inc., DoorDash Inc. and Grab Holdings Ltd. and share prices fell, regulatory filings show. The Vision Fund was being systematically unwound, with Son selling anything that could generate cash.

Even Alibaba, the crown jewel that had defined SoftBank's investment success, wasn't sacred. By 2024, SoftBank had sold most of its Alibaba stake, ending a 24-year relationship that had generated returns of over 400x. The company that had once defined itself by its portfolio was now just a holding company with a chip designer and a lot of cash.

The pivot to AI wasn't just rhetoric. In January 2025, it was announced that Softbank, along with OpenAI, MGX, and Oracle, would launch what was announced to be an artificial intelligence infrastructure system in conjunction with the U.S. The Stargate project, unveiled by President Trump, called for hundreds of billions of dollars of investment into AI infrastructure. Son was betting that owning the infrastructure of AI—the chips, the data centers, the networks—was more valuable than investing in AI applications. The Stargate announcement in January 2025 represented Son's most audacious pivot yet. The Stargate Project is a new company which intends to invest $500 billion over the next four years building new AI infrastructure for OpenAI in the United States. We will begin deploying $100 billion immediately. The initial equity funders in Stargate are SoftBank, OpenAI, Oracle, and MGX. SoftBank and OpenAI are the lead partners for Stargate, with SoftBank having financial responsibility and OpenAI having operational responsibility. Masayoshi Son will be the chairman.

The scale was breathtaking even by Son's standards. Three top tech firms on Tuesday announced that they will create a new company, called Stargate, to grow artificial intelligence infrastructure in the United States. OpenAI CEO Sam Altman, SoftBank CEO Masayoshi Son and Oracle Chairman Larry Ellison appeared at the White House Tuesday afternoon alongside President Donald Trump to announce the company, which Trump called the "largest AI infrastructure project in history."

But questions immediately emerged about the financing. "They don't actually have the money," Musk wrote on X, the social media app he owns. "SoftBank has well under $10B secured. I have that on good authority." The Wall Street Journal reported that SoftBank, the main financial backer for Stargate, could collect only a 10% equity funding and the remaining would come from debt or debt-like fundings.

The Vision Fund, meanwhile, had become a shadow of its former self. The one-time tech kingmaker is now a shadow of its former self, having laid off more than a hundred staff and slowed new investments to a fraction of its past pace. Kirk Boodry, an analyst at Astris Advisory, estimates the two Vision Funds sold at least $6 billion worth of their holdings in the fiscal year ended March. The first Vision Fund has sold at least $5 billion since Sept. 24, he said. "The Vision Fund has been a prolific seller since September," he said. "A growing cash pile could point to a deeper pivot to generative AI."

The lessons learned were painful but clear. While Son long teased the possibility of a series of Vision Funds launched every two to three years, the prospect of a third Vision Fund—let alone a fourth—no longer comes up, said the people. Instead, the fund's reduced staff are mostly caretakers. They are concentrating on locking in investment gains and reversing any losses.

Son's new philosophy was fundamentally different from the Vision Fund era. Instead of betting on hundreds of companies hoping a few would succeed, he was betting on the infrastructure that all AI companies would need. Instead of funding growth at any cost, he was investing in companies with actual technology moats. Instead of chasing unicorns, he was building utilities.

The ARM position exemplified this shift. The chip designer's market value has soared to around $106 billion since its market debut last year, making SoftBank's 90% holding worth more than all of SoftBank. That's helped lift SoftBank's cash pile to ¥6.2 trillion, up from ¥4.6 trillion at the end of 2021. Rather than using this windfall to fund more venture investments, Son was hoarding cash for strategic bets on AI infrastructure.

Looking back, the Vision Fund reckoning represented the end of an era—not just for SoftBank but for venture capital as a whole. The fund that had defined the unicorn era with its massive checks and growth-at-all-costs philosophy had become a cautionary tale about the dangers of too much capital and too little discipline. Five years after Masayoshi Son's $100 billion fund entered the financial world to much fanfare, Softbank's venture firm was crumbling and on the verge of collapse. Its large venture vehicles struggled badly, performing in the bottom of the asset class, and many of Son's closest associates in the effort had departed from the company.

The pivot to AI infrastructure wasn't just a new investment strategy—it was an admission that the Vision Fund model had failed. Son was no longer trying to pick winners in the innovation economy. He was trying to own the picks and shovels. Whether this new strategy would succeed where the Vision Fund failed remained to be seen. But one thing was clear: the era of SoftBank as venture capital's biggest player was over.

↑ Back to Top

XII. Playbook: Investment Philosophy & Lessons

After four decades of dealmaking, Masayoshi Son's investment philosophy can be distilled into a single image: the slide he shows at every major presentation, depicting a 300-year timeline with humanity ascending toward "the singularity." This isn't hyperbole or marketing—it's the operating system that drives every SoftBank decision. Understanding this worldview is essential to understanding both SoftBank's spectacular successes and catastrophic failures.

The "Vision" Approach: Betting on Paradigm Shifts

Son doesn't invest in companies; he invests in inflection points. We attribute the rise of Softbank to its ability to leverage asymmetries in three types of information. These were the value-chain asymmetry between software developers and retailers; the financial asymmetry between practices in the US and Japan; and, finally, the technology asymmetry between computer products introduced first in the US and later in Japan. But these tactical advantages mask a deeper strategic insight: Son consistently identifies when a technology is about to shift from experimental to essential.

The pattern is remarkably consistent across decades. In the 1980s, he saw that personal computers would transform from hobbyist curiosities to business necessities. In the 1990s, he recognized the internet would become infrastructure, not entertainment. In 2000, he bet that China's e-commerce would leapfrog physical retail. In 2016, he wagered that ARM's chip designs would power the IoT revolution. Each bet was made before the trend was obvious but after the technology was proven—what venture capitalists call the "sweet spot" but executed at a scale that makes traditional VCs look like they're playing with pocket change.

Capital as a Weapon: The Pros and Cons of Massive Funds

Son's most controversial innovation wasn't identifying trends but weaponizing capital. Traditional venture capital assumes scarcity—limited funds mean careful selection and gradual deployment. Son inverted this: with unlimited capital, you could guarantee winners by eliminating competition. Give Uber $9 billion, and it doesn't matter if the unit economics work—competitors simply can't match the subsidy war.

This strategy worked brilliantly when SoftBank was the only player with this scale of capital. The Vision Fund could single-handedly set valuations for entire sectors. But the strategy had three fatal flaws. First, it attracted the wrong entrepreneurs—those skilled at raising money rather than building businesses. Second, it created artificial markets where companies competed on fundraising rather than fundamentals. Third, and most critically, it assumed capital alone could overcome structural business model problems.

WeWork exemplified all three failures. Adam Neumann was a fundraising savant but an operational disaster. The co-working market became distorted by WeWork's capital-fueled expansion. And no amount of money could fix the fundamental mismatch between long-term lease obligations and short-term revenue contracts.

The Importance of Timing in Technology Cycles

Son's greatest skill isn't picking technologies but timing them. He enters markets after the technology works but before it scales—what he calls the "critical mass point." Yahoo! Japan launched just as Japanese internet penetration was exploding. The Alibaba investment came as Chinese e-commerce was transitioning from experiment to standard. The Vision Fund launched at peak unicorn mania, allowing Son to deploy capital when founders were most receptive to massive checks.

But timing cuts both ways. The Sprint acquisition came too late—the U.S. telecom market was already consolidated. The Vision Fund's growth-at-all-costs philosophy peaked just as markets started demanding profitability. The ARM acquisition, initially questioned, now looks prescient as AI drives demand for specialized chips. Success in Son's model isn't just about being early—it's about being perfectly calibrated to the adoption curve.

Managing a Conglomerate vs. Pure-Play Investment Firm

SoftBank's structure—part operating company, part investment firm—creates unique advantages and challenges. The operating businesses (telecom, ARM) provide stable cash flows that fund speculative investments. The investment wins (Alibaba) provide capital for acquisitions. This circular flow of capital allows SoftBank to be patient in ways pure investment firms cannot.

But the structure also creates conflicts. Operating businesses require focus, discipline, and gradual improvement. Investment activities demand risk-taking, speed, and acceptance of failure. The cultures clash. Telecom executives who spend years optimizing network efficiency watch Son write billion-dollar checks in minutes. Investment teams chasing the next unicorn clash with ARM engineers methodically improving chip designs.

Son has never resolved this tension. Instead, he compartmentalizes ruthlessly. The operating businesses run independently with minimal interference. The investment activities report directly to Son. This works when both sides succeed but creates pressure when either struggles. The 2020 crisis forced Son to choose between supporting the Vision Fund or protecting the operating businesses. He chose survival over strategy, selling assets to maintain liquidity.

Key Mistakes: Overriding Diligence, Founder Worship, Valuation Discipline

The Vision Fund era exposed three systematic errors in Son's approach. First, he consistently overrode due diligence when he liked a founder. The WeWork investment proceeded despite internal objections about governance, business model, and valuation. The Wirecard investment ignored obvious fraud signals. Zume Pizza, Katerra, Greensill—the list of diligence failures is long and expensive.

Second, Son's founder worship created blind spots. He saw himself in entrepreneurs like Adam Neumann and Nikola's Trevor Milton—visionary outsiders who would transform industries through force of will. This projection prevented him from seeing obvious flaws. Neumann's self-dealing, Milton's fraud, the Theranos-style promises of multiple portfolio companies—all were missed because Son was evaluating founders as fellow visionaries rather than as managers of capital.

Third, and most critically, Son abandoned valuation discipline during the Vision Fund era. His famous "gut-based" investing worked when he was writing $20 million checks. But when the checks reached $1 billion, gut instinct wasn't enough. The Vision Fund paid prices that assumed perfect execution, no competition, and ever-expanding markets. When any of these assumptions broke, the investments collapsed.

The Asymmetric Bet Philosophy

Despite the failures, Son's core insight remains valid: in technology markets, the upside of being right dwarfs the downside of being wrong. One Alibaba pays for a hundred WeWorks. This asymmetry justifies aggressive betting—if you can survive the losses. Son structures SoftBank to maximize this asymmetry. The holding company structure limits contagion from failed bets. The portfolio approach ensures multiple shots on goal. The long-term orientation allows riding out cycles.

But this philosophy only works with certain types of bets. Infrastructure investments (telecom, chips, data centers) have limited upside but predictable returns. Platform investments (Alibaba, Yahoo! Japan) have massive upside if they achieve monopoly. The Vision Fund confused the two, treating WeWork (infrastructure) like Alibaba (platform). The result was catastrophic because infrastructure businesses can't generate platform returns.

Evolution and Adaptation

Son's greatest strength may be his ability to admit error and pivot. After the dot-com crash, he shifted from pure internet investing to telecom operations. After the WeWork debacle, he abandoned the Vision Fund model for strategic AI investments. Each failure produces a course correction, though usually after enormous losses.

The current AI infrastructure strategy represents Son's most conservative approach yet. Instead of betting on winners in the AI race, he's providing the computing power all AI companies need. It's the equivalent of selling shovels during a gold rush—less upside but more certainty. Whether this conservatism is wisdom or exhaustion remains to be seen.

The Ultimate Lesson

The SoftBank story teaches that in technology investing, three things matter more than everything else: timing, capital, and conviction. Get all three right, and you create Alibaba—one of the greatest investment returns in history. Get even one wrong, and you create WeWork—one of the most spectacular destructions of value ever seen. Son has been on both extremes, sometimes simultaneously.

The playbook, ultimately, isn't about specific tactics or strategies. It's about a worldview that sees technology as destiny and capital as the means to accelerate that destiny. Whether you agree with this worldview determines whether you see Son as a visionary or a gambler. The truth, as always with SoftBank, is probably both.

↑ Back to Top

XIII. Bear vs. Bull Case

Bear Case: The Skeptic's View

The bear case for SoftBank writes itself in red ink across two decades of destroyed value. Start with the Vision Fund's track record: After generating initial paper gains, the funds have produced cumulative losses that would bankrupt most investment firms. The $100 billion Vision Fund, after eight years, is worth less than what investors committed—a devastating indictment of Son's investment philosophy. As one Gulf investor noted, "You can't catch up for that lost period" when discussing why neither Saudi Arabia's PIF nor Abu Dhabi's Mubadala reinvested in Vision Fund 2.

The concentration risk in Chinese technology remains alarming despite the Alibaba exit. SoftBank still holds significant positions in companies vulnerable to U.S.-China tensions, regulatory crackdowns, and geopolitical uncertainty. Every escalation in the tech cold war threatens billions in portfolio value. The Chinese government's swift neutering of Ant Financial's IPO demonstrated how quickly fortunes can reverse.

Governance concerns persist despite cosmetic improvements. Son, at 67, shows no signs of succession planning. The board remains dominated by Son loyalists. The company's structure—a complex web of holding companies, subsidiaries, and investment vehicles—makes true oversight nearly impossible. When Son eventually steps down, SoftBank could face an existential crisis as no clear successor possesses his unique combination of vision, relationships, and risk tolerance.

Competition from more disciplined investors has intensified. While SoftBank was burning billions on WeWork, firms like Sequoia, Andreessen Horowitz, and Tiger Global refined their models. These firms now deploy capital at scale but with better governance, clearer strategies, and stronger track records. The era of SoftBank's capital advantage has ended.

The debt burden, while manageable, limits flexibility. SoftBank carries approximately $150 billion in net debt across its various entities. Rising interest rates increase servicing costs. Any significant portfolio shock could trigger a liquidity crisis. The 2020 emergency asset sale program demonstrated how quickly SoftBank can find itself cornered.

Most fundamentally, the Vision Fund proved that Son's investment philosophy—massive capital deployment to create monopolies—doesn't work in competitive markets. The assumption that capital alone can guarantee success has been thoroughly debunked. Without this edge, SoftBank is just another investment firm with a spotty track record and excessive overhead.

Bull Case: The Believer's Perspective

The bull case starts with ARM, the hidden gem that validates Son's strategic vision. ARM's architecture powers 95% of smartphones and increasingly dominates AI inference, edge computing, and IoT devices. With a market value exceeding $150 billion and SoftBank owning 90%, this single asset nearly justifies SoftBank's entire market capitalization. As AI pushes computing to the edge, ARM's value could multiply.

The early positioning in AI infrastructure through Stargate could define the next decade. The $500 billion commitment to AI data centers, while ambitious, positions SoftBank at the center of the AI revolution. Unlike the Vision Fund's spray-and-pray approach, this focused bet on infrastructure provides predictable returns from an inevitable trend. As Son noted, SoftBank is selling shovels in the AI gold rush—a proven strategy for wealth creation.

The T-Mobile stake appreciation demonstrates the value of patient capital. What looked like a failed Sprint acquisition transformed into an 8.1x return through financial engineering and market appreciation. SoftBank still holds valuable stakes that could be similarly revalued. The ability to hold positions for decades, unlike traditional funds with fixed lifespans, creates unique advantages.

Son's track record of one transformative bet per decade remains intact. Yahoo! Japan in the 1990s, Alibaba in the 2000s, the mobile transformation in the 2010s—each generated returns that justified all other failures. The AI infrastructure bet could be the 2020s equivalent. History suggests betting against Son's ability to identify paradigm shifts is dangerous.

The balance sheet has never been stronger. With ¥6.2 trillion in cash and sellable assets, SoftBank can weather any storm. The company has learned from 2020, maintaining larger liquidity buffers and reducing leverage. This financial strength allows patient investment in long-term opportunities without market pressure.

Finally, the SoftBank ecosystem creates unique advantages. Portfolio companies can collaborate, share resources, and access SoftBank's global network. ARM's technology enables SoftBank's AI investments. The telecom operations provide steady cash flow. This interconnected model, while complex, generates synergies pure investment firms cannot replicate.

The Balanced View

The truth lies between extremes. SoftBank is neither the visionary force Son promotes nor the disaster bears describe. It's a complex conglomerate with spectacular successes and failures, struggling to evolve beyond its founder's personality.

The Vision Fund's failure was real but not fatal. The losses, while massive, were absorbed without destroying the company. The lessons learned—about governance, diligence, and valuation discipline—appear to be genuine. The pivot from venture investing to infrastructure plays to SoftBank's strengths while avoiding its weaknesses.

The succession issue looms but isn't immediate. Son shows no signs of slowing down, and the simplified structure (focusing on ARM and AI infrastructure) is more manageable for eventual successors. The cult of personality that enabled both great successes and failures will eventually end, potentially creating a more stable if less spectacular company.

Market conditions favor SoftBank's evolution. The era of free money that enabled Vision Fund excess has ended, but the AI revolution creates new opportunities. SoftBank's combination of patient capital, strategic assets, and global reach positions it well for this transition—if it can maintain discipline.

The investment case depends on time horizon and risk tolerance. Short-term investors should stay away—SoftBank's volatility and complexity create constant uncertainty. Long-term investors might find value, particularly if they believe in the AI transformation and Son's ability to capture value from it.

Ultimately, SoftBank remains what it's always been: a leveraged bet on technological transformation, embodied in one man's vision. Whether that's an opportunity or a warning depends entirely on your faith in both the transformation and the man. The bear case is compelling, the bull case is plausible, but the reality is that SoftBank defies conventional analysis. It's not a company you analyze—it's one you either believe in or you don't.

↑ Back to Top

XIV. Epilogue: The Future of SoftBank

Masayoshi Son stood before the Stargate announcement at the White House in January 2025, flanked by Sam Altman and Larry Ellison, and allowed himself a small smile. At 67, after four decades of building SoftBank, he was starting over again. The Vision Fund era was behind him. The age of AI infrastructure had begun. "Now I came back with $500 billion," he quipped to President Trump, who had jokingly asked him to double his previous $100 billion commitment. Classic Son—when challenged to think bigger, he quintupled down.

The AI infrastructure play represents both continuity and change in SoftBank's evolution. The continuity is obvious: another massive bet on a technological paradigm shift, executed at a scale that makes conventional investors queasy. The change is subtle but significant: instead of betting on hundreds of companies hoping to find winners, Son is building the infrastructure all AI companies will need. It's the difference between gambling on gold miners and owning the land they mine on.

Vision Fund's evolution tells its own story of learning and adaptation. The first fund's spectacular failures—WeWork, Wirecard, Greensill—taught expensive lessons about due diligence, governance, and the limits of capital as strategy. The second fund, smaller and entirely self-funded, reflects a chastened approach. There will be no Vision Fund 3. The era of writing billion-dollar checks to any entrepreneur with charisma and a pitch deck has ended.

The lessons learned are evident in current strategy. Instead of funding companies that lose money on every transaction while hoping for scale, SoftBank now focuses on businesses with actual moats. ARM's processor architecture, AI data centers, strategic stakes in platform companies—these are defensible positions that compound value over time. The spray-and-pray approach has given way to concentrated bets on infrastructure.

Son's legacy will be complex and contradictory. He democratized venture capital by proving that massive funds could generate returns, then proved that massive funds could destroy value just as quickly. He identified every major technology shift of the past forty years, but frequently botched the execution. He made one of the greatest investment returns in history with Alibaba, then squandered much of it on WeWork. He's simultaneously the best and worst investor of his generation.

The succession question remains unanswered and perhaps unanswerable. Son has no obvious heir, no protégé groomed to take over. The executives who might have succeeded him—Nikesh Arora, Marcelo Claure, Rajeev Misra—have all departed. The company's future post-Son is genuinely uncertain. Will it become a conventional conglomerate, carefully managing its assets? Will it break up, with ARM spun off and the investments liquidated? Or will some unknown successor emerge with their own vision for SoftBank's next chapter?

What comes next depends partly on macroeconomic conditions beyond anyone's control. If AI delivers the productivity revolution optimists predict, SoftBank's infrastructure investments could generate enormous returns. If AI proves overhyped, like many technologies before it, SoftBank could face another reckoning. The company's leveraged structure amplifies both outcomes.

Final Thoughts on SoftBank's Impact on Global Tech Investing

SoftBank's impact on global technology investing is undeniable and irreversible. Before the Vision Fund, venture capital was a cottage industry of small funds making careful bets. After it, venture capital became an arms race of ever-larger funds chasing ever-higher valuations. The "blitzscaling" mentality—grow at any cost, winner takes all—was enabled and encouraged by SoftBank's capital.

This transformation had both positive and negative effects. Positively, it accelerated innovation by giving entrepreneurs resources to think bigger and move faster. Companies that might have taken decades to reach global scale did so in years. Technologies that might have developed gradually were turbocharged. The smartphone revolution, the gig economy, the sharing economy—all were accelerated by SoftBank capital.

Negatively, it distorted markets and destroyed value. Companies optimized for fundraising rather than building sustainable businesses. Valuations disconnected from fundamentals. The discipline that comes from capital scarcity disappeared. The result was a generation of companies that could only survive with continuous capital injections—corporate zombies kept alive by financial engineering.

SoftBank also changed how we think about technology investing itself. The traditional venture capital model assumed multiple small bets with a few massive winners. Son inverted this: make massive bets assuming most would win. When it worked (Alibaba), the returns were extraordinary. When it failed (WeWork), the losses were equally spectacular. This high-variance approach redefined risk in technology investing.

The company's influence extends beyond pure economics. SoftBank proved that Asian capital could reshape Silicon Valley, ending the era of American venture capital hegemony. It demonstrated that sovereign wealth funds could be active technology investors, not just passive allocators. It showed that one visionary with enough capital could bend entire industries to their will—for better or worse.

Looking forward, SoftBank's next chapter will help determine whether the AI revolution follows the internet's path or charts something entirely new. If Stargate succeeds in building AI infrastructure at unprecedented scale, it could accelerate AI deployment by years. If it fails, it might prove that even in technology, there are limits to what capital can achieve.

The ultimate judgment on SoftBank depends on timeframe. In the short term, it's been a destroyer of value, burning hundreds of billions in pursuit of Son's vision. In the long term, it might be remembered as the catalyst that transformed how humanity deploys capital to accelerate technological progress. The final verdict won't be clear for decades.

What is clear is that SoftBank, and Masayoshi Son, have left an indelible mark on how we fund, build, and think about technology companies. Every massive funding round, every blitzscaling strategy, every sovereign wealth fund writing venture checks—all bear SoftBank's influence. The company might not survive in its current form, Son will eventually step down, but the SoftBank model has permanently altered the technology landscape.

As Son himself might say, looking at his 300-year timeline, we're still in the early innings. The information revolution he's spent his life funding and fostering has barely begun. Whether SoftBank continues to play a central role in that revolution, or becomes merely a historical curiosity—a cautionary tale of ambition exceeding execution—remains to be written.

The only certainty is that it won't be boring. With Son at the helm, SoftBank will continue swinging for the fences, making massive bets on transformative technologies, and either spectacular succeeding or spectacularly failing—probably both, simultaneously. That's the SoftBank way: always ambitious, sometimes successful, never dull. In a world of increasingly homogenized corporations, that itself might be Son's greatest legacy.

↑ Back to Top

XV. Recent News

The January 2025 Stargate announcement represents SoftBank's most dramatic pivot yet. Standing alongside President Trump, Son committed to deploying $100 billion immediately as part of a $500 billion AI infrastructure project over four years. The project, structured as a new company with SoftBank holding financial responsibility and OpenAI managing operations, signals Son's belief that AI infrastructure—not AI applications—will define the next decade's winners.

The timing is notable. Just months after completing SoftBank's exit from Alibaba—ending a 24-year relationship that generated over 400x returns—Son is recycling that capital into physical infrastructure. The contrast with the Vision Fund era couldn't be starker: instead of funding software companies with nebulous paths to profitability, SoftBank is building data centers with contracted revenue from the world's leading AI companies.

ARM's performance continues to validate Son's semiconductor thesis. The chip designer's stock has surged 124% year-to-date through July 2024, driven by increasing adoption in AI inference and edge computing. With major cloud providers designing custom ARM-based chips and Apple's successful M-series processors proving ARM's viability beyond mobile, the architecture Son acquired for $32 billion now underpins computing's future.

The Vision Fund unwinding accelerates. Recent filings show systematic selling across the portfolio, with over $6 billion in disposals in the latest fiscal year. The fund that once wrote billion-dollar checks monthly now functions primarily as a liquidation vehicle, harvesting remaining value while avoiding new commitments. The staff reductions—from 500 employees at peak to fewer than 100 today—reflect this reality.

Financial results show a company in transition. SoftBank's cash position has strengthened to ¥6.2 trillion, up from ¥4.6 trillion in 2021, providing flexibility for the AI infrastructure pivot. The debt reduction program continues, with net debt declining even as the company commits massive capital to Stargate. This deleveraging while investing represents a maturation from Son's historical approach of borrowing aggressively to fund bets.

Regulatory scrutiny intensifies around the Stargate project. Questions about national security implications of foreign investment in critical AI infrastructure, the concentration of power in AI development, and the environmental impact of massive data centers all require navigation. Unlike the Vision Fund era's purely financial engineering, Stargate involves physical infrastructure that intersects with government interests.

The competitive landscape has shifted dramatically since SoftBank's Vision Fund heyday. Microsoft's $10 billion OpenAI partnership, Amazon's Anthropic investment, and Google's internal AI development all represent alternative models to SoftBank's infrastructure play. The question isn't whether AI will transform computing but whether owning infrastructure provides better returns than owning applications.

Market reception remains skeptical but improving. SoftBank's stock has risen 78% year-to-date, largely driven by ARM's appreciation, but still trades at a significant discount to its net asset value. This persistent "SoftBank discount" reflects continued uncertainty about Son's latest pivot and concerns about execution risk in the Stargate project.

↑ Back to Top

XVI. Links & Resources

Essential Long-Form Reads:

"The Rise and Fall of the Vision Fund" - The Information's comprehensive investigation into the Vision Fund's investment process, internal conflicts, and spectacular failures provides unparalleled insight into how $100 billion was deployed and often destroyed.

"Masayoshi Son's 300-Year Plan" - Bloomberg Businessweek's profile explores Son's philosophical approach to investing and his belief in the technological singularity that drives every major SoftBank decision.

"The WeWork Debacle: An Autopsy" - The Wall Street Journal's definitive account of how SoftBank turned a co-working company into a $47 billion valuation before watching it collapse to near bankruptcy.

"Alibaba and SoftBank: The Greatest Venture Return Ever" - Fortune's analysis of how a $20 million investment became worth over $60 billion, fundamentally changing both companies' trajectories.

"Inside the Vision Fund" - Financial Times' series examining the fund's structure, governance failures, and the culture clash between Silicon Valley disruption and Gulf sovereign wealth.

Books for Deeper Understanding:

"Aiming High" by Atsuo Inoue - The only authorized biography of Masayoshi Son, providing crucial context on his Korean-Japanese background and the discrimination that shaped his worldview.

"The Power Law" by Sebastian Mallaby - While not exclusively about SoftBank, this venture capital history positions the Vision Fund within the broader evolution of technology investing.

"Billion Dollar Loser" by Reeves Wiedeman - The definitive account of WeWork's rise and fall, with extensive detail on SoftBank's role in enabling Adam Neumann's excess.

Regulatory Filings and Primary Sources:

SoftBank Group's investor relations portal provides quarterly earnings presentations where Son personally explains strategy shifts, portfolio performance, and future vision with his characteristic slides.

The Vision Fund's limited partner reports, while not public, have occasionally leaked, providing rare insight into how the world's largest tech fund presents its performance to sovereign wealth investors.

Academic and Industry Analysis:

"SoftBank: A Case Study in Conglomerate Complexity" - Harvard Business School's analysis of SoftBank's structure, governance, and the challenges of managing a company that defies traditional categorization.

"The SoftBank Effect on Venture Capital Markets" - Stanford Graduate School of Business research on how the Vision Fund's capital deployment changed startup valuations, competitive dynamics, and exit strategies globally.

---

↑ Back to Top

Conclusion: The Paradox of Masayoshi Son

SoftBank defies conventional analysis because it has never been a conventional company. It's a reflection of one man's unshakeable belief that technology represents humanity's destiny, and that capital—deployed at massive scale—can accelerate that destiny. Whether this makes Masayoshi Son a visionary or a gambler isn't the right question. He's both, simultaneously, and that paradox is SoftBank's essential nature.

The company's journey from software distributor to AI infrastructure investor encapsulates the entire digital revolution. Each phase—PC software, internet, mobile, AI—brought spectacular successes and failures, often simultaneously. The Alibaba investment remains one of venture capital's greatest returns. The WeWork debacle remains one of its most spectacular destructions of value. Both happened at the same company, driven by the same philosophy, executed by the same person.

What distinguishes SoftBank isn't its track record—which is simultaneously brilliant and terrible—but its impact on how technology gets funded globally. Before Son, venture capital was a cottage industry of careful bets and patient capital. After him, it became an arms race of massive funds and blitzscaling strategies. This transformation has accelerated innovation while also creating distortions that may take decades to fully understand.

The current pivot to AI infrastructure suggests Son has learned from the Vision Fund's failures without abandoning his core belief in transformative bets. Building data centers and owning chip architectures is less exciting than funding unicorns, but it's also less likely to implode. Whether this represents wisdom or exhaustion will only be clear in retrospect.

The succession question looms larger with each passing year. Son, at 67, shows no signs of slowing down, but SoftBank's dependence on his vision becomes more problematic over time. The company has proven it can survive massive losses, market crashes, and portfolio implosions. Whether it can survive the departure of its founder remains the ultimate test.

For investors, SoftBank represents a unique proposition: exposure to technological transformation at a scale impossible to replicate individually, but with volatility and complexity that challenge traditional analysis. It's not a company you own for stable returns or predictable growth. It's a company you own if you believe, as Son does, that technology will fundamentally reshape human civilization and that owning the infrastructure of that transformation will generate extraordinary returns.

The bear case—Vision Fund losses, governance concerns, succession risks—is compelling. The bull case—ARM's dominance, AI infrastructure positioning, Son's track record of identifying paradigm shifts—is equally strong. But perhaps the most accurate view is that SoftBank exists outside traditional investment frameworks. It's not a value investment or a growth investment. It's a bet on transformation itself.

Looking ahead, SoftBank's next chapter will be defined by the success or failure of the Stargate project and the broader AI infrastructure bet. If AI delivers the productivity revolution many predict, SoftBank's positioning could generate returns that dwarf even Alibaba. If AI disappoints, the company faces another painful reckoning. There's no middle ground in Son's worldview—only revolutionary success or spectacular failure.

The irony is that even SoftBank's failures have been productive. The Vision Fund's excess forced the venture capital industry to reconsider the relationship between capital and value creation. WeWork's implosion taught important lessons about governance and business model sustainability. Even Sprint, Son's longest-running struggle, eventually generated substantial returns through financial engineering.

This pattern—failure leading to learning leading to evolution—defines SoftBank's resilience. The company that started distributing PC software has survived and thrived through multiple technological revolutions not by avoiding mistakes but by surviving them and adapting. Each crisis produces a strategic pivot. Each pivot positions SoftBank for the next wave of transformation.

Critics argue that SoftBank is just a vehicle for Son's gambling addiction, dressed up in technological rhetoric. Supporters see it as the purest expression of technological optimism, backed by capital at a scale that can actually move markets. Both views contain truth. SoftBank is simultaneously reckless and visionary, destructive and creative, a cautionary tale and an inspiration.

The ultimate verdict on SoftBank won't be clear for decades. Son's 300-year vision may seem absurd, but his four-decade track record of identifying and capitalizing on technological shifts suggests dismissing him is dangerous. The company that brought the internet to Japan, funded China's e-commerce revolution, and reshaped venture capital globally isn't finished surprising us.

In the end, SoftBank's story is about the power and peril of conviction. Son's unwavering belief in technological transformation has created enormous value and destroyed enormous value, sometimes simultaneously. Whether that conviction ultimately proves prophetic or delusional, it has undeniably shaped how humanity funds and develops technology. In a world increasingly defined by technological change, that impact may be Son's greatest legacy—regardless of SoftBank's final fate.

The company enters its fifth decade with the same combination of massive ambition and existential risk that has defined it from the beginning. The AI infrastructure bet represents another roll of the dice, another test of Son's vision against market reality. History suggests betting against him is dangerous. History also suggests betting on him is terrifying. That paradox—the source of both SoftBank's greatest triumphs and most spectacular failures—remains unresolved. And as long as Masayoshi Son remains at the helm, it probably always will be.

The story continues with recent developments that suggest SoftBank's evolution is far from complete. In November 2024, the company announced a ¥500 billion buyback program, its largest since the 2020 crisis, signaling confidence in the AI pivot while acknowledging the persistent discount between SoftBank's market value and its assets. The buyback represents a shift from empire-building to value realization—a tacit admission that markets don't trust Son's capital allocation but might trust the underlying assets.

The telecommunications business, often forgotten amid the investment drama, continues generating steady cash flows. SoftBank's Japanese mobile operations, built on the bones of Vodafone's failure, now serve over 50 million subscribers with industry-leading 5G coverage. This operational excellence in telecom stands in stark contrast to the chaos of the Vision Fund, suggesting SoftBank can execute when it focuses on actual operations rather than financial engineering.

Recent board changes hint at potential succession planning. The appointment of CFO Yoshimitsu Goto to the board in 2023, along with the elevation of several younger executives, suggests Son is finally building a leadership bench. These moves, while incremental, represent the first serious attempt to prepare SoftBank for a post-Son era.

The regulatory environment has shifted dramatically since SoftBank's freewheeling Vision Fund days. The Committee on Foreign Investment in the United States (CFIUS) now scrutinizes any significant technology investment. The European Union's Digital Markets Act imposes new constraints on platform companies. China's tech crackdown eliminated entire categories of investment. The Stargate project must navigate this more complex landscape, where geopolitics matters as much as economics.

Environmental, social, and governance (ESG) considerations, largely ignored during the Vision Fund era, now influence investment decisions. The massive energy consumption of AI data centers raises sustainability questions that SoftBank must address. The company's response—committing to renewable energy for Stargate facilities—represents another evolution from the growth-at-any-cost mentality.

The Competitive Landscape Transforms

The venture capital ecosystem SoftBank disrupted has adapted and evolved. Tiger Global, once dismissed as a SoftBank imitator, has developed a more disciplined model of rapid deployment without the governance failures. Andreessen Horowitz raised $35 billion across multiple funds, approaching Vision Fund scale but with better diversification. Even traditional firms like Sequoia have launched billion-dollar funds, eroding SoftBank's capital advantage.

Corporate venture arms have emerged as serious competitors. Microsoft's strategic investments in OpenAI and other AI companies combine capital with distribution and technical resources SoftBank can't match. Google's internal AI development, funded by search profits, doesn't need external capital at all. These tech giants can match SoftBank's financial resources while offering strategic advantages Son can't replicate.

Sovereign wealth funds, once passive limited partners in the Vision Fund, now invest directly. Saudi Arabia's PIF, Singapore's Temasek, and Abu Dhabi's Mubadala have built internal teams and make their own technology bets. They learned from the Vision Fund experience—both what to do and what not to do. The student has become the competitor.

The China Question

SoftBank's relationship with China has transformed from opportunity to liability. The Alibaba exit, while financially successful, represents the end of an era when Son could freely navigate between East and West. The new reality requires choosing sides in the technology cold war. The Stargate project's emphasis on U.S. infrastructure and partnership with American companies signals that choice.

Yet China remains entangled in SoftBank's portfolio through numerous Vision Fund investments. ByteDance, Didi, and other Chinese unicorns face uncertain futures as regulators on both sides impose restrictions. These positions, once viewed as crown jewels, now represent stranded assets that may never achieve liquidity.

The geopolitical shift affects ARM as well. The chip designer's technology is crucial for both American and Chinese companies, creating diplomatic complexity. Every licensing decision becomes a political statement. The neutrality that allowed ARM to dominate globally now seems impossible to maintain.

Technology Trends and SoftBank's Position

The AI revolution differs fundamentally from previous technology waves SoftBank has ridden. The internet and mobile revolutions democratized access to information and communication. AI concentrates power in the hands of those with data, computing resources, and specialized talent. This concentration plays to SoftBank's strengths—capital can buy computing power—but also exposes weaknesses in talent acquisition and retention.

Quantum computing looms as the next paradigm shift. SoftBank has made small investments through the Vision Fund but lacks the deep technical expertise to evaluate quantum opportunities. Unlike software investments where Son could rely on intuition, quantum requires understanding physics at a level that few investors possess.

The metaverse, briefly hyped as the next platform, has largely disappointed. SoftBank's limited exposure—mainly through gaming investments—looks prescient in hindsight. But this also reflects a new conservatism, missing opportunities that the younger Son might have pursued despite skepticism.

Biotechnology represents an enormous opportunity that SoftBank has largely ignored. The convergence of AI and biology could transform medicine, agriculture, and materials science. Son's exclusive focus on information technology may cause him to miss the bio-revolution, just as his early focus on software initially caused him to miss hardware opportunities.

Financial Engineering in the Modern Era

SoftBank's financial complexity has reached new heights with the Stargate structure. The use of special purpose vehicles, preferred equity structures, and complex derivatives makes the company nearly impossible for ordinary investors to value. This opacity might be intentional—it allows Son maximum flexibility while minimizing scrutiny.

The relationship between SoftBank Group, the Vision Funds, and various subsidiaries creates potential conflicts of interest that regulators increasingly scrutinize. The ARM IPO, structured to maintain SoftBank's control while accessing public markets, exemplifies this complexity. Every transaction requires armies of lawyers and bankers to navigate regulatory requirements across multiple jurisdictions.

Currency exposure adds another layer of complexity. With assets denominated in dollars, yen, yuan, and euros, SoftBank functions as a massive currency bet alongside its technology investments. The yen's weakness has flattered recent results, but currency moves could just as easily destroy value.

The Human Capital Challenge

SoftBank's inability to retain top talent remains a critical weakness. The departures of Nikesh Arora, Marcelo Claure, and Rajeev Misra represent more than just executive turnover—they reflect a culture where only Son's vision matters. This creates a competency gap that becomes more problematic as the company's complexity increases.

The Vision Fund's London office, once bustling with hundreds of investment professionals, now operates with skeleton staff. The expertise in evaluating technology investments, built over years, has dispersed to competitors. Rebuilding this capability would take years that SoftBank may not have.

Younger employees increasingly question SoftBank's culture of extreme risk-taking and top-down decision-making. The new generation of technology professionals has options—they can join established tech giants, well-funded startups, or start their own companies. SoftBank's appeal as an employer has diminished alongside its reputation as an investor.

The Path Forward

SoftBank faces three potential paths, each with distinct implications for stakeholders. The first path involves doubling down on AI infrastructure, essentially becoming a technology utility company. This would provide stable returns but abandon Son's vision of revolutionary transformation. The ARM franchise and Stargate project could anchor this strategy, providing predictable cash flows from the AI revolution's infrastructure needs.

The second path involves breaking up the conglomerate. ARM could be fully spun off, capturing its full value for shareholders. The telecom operations could merge with competitors. The Vision Fund investments could be liquidated or distributed to shareholders. This financial engineering would likely create immediate value but end SoftBank as a coherent entity.

The third path, and the most likely while Son remains in charge, involves continued evolution while maintaining the core vision. This means smaller, more focused bets on transformative technologies. It means accepting lower returns in exchange for lower risk. It means preparing for succession while Son remains active. This middle path satisfies no one completely but might be the only sustainable option.

Lessons for the Industry

SoftBank's journey offers crucial lessons for investors, entrepreneurs, and regulators. The first lesson concerns the relationship between capital and innovation. While capital can accelerate development, it can't create innovation where none exists. WeWork had unlimited capital but couldn't transform a fundamentally flawed business model. Alibaba had limited capital but solved a real problem for Chinese businesses.

The second lesson involves governance and accountability. The Vision Fund's failures stemmed partly from Son's unchecked power to override investment committees and ignore red flags. Even visionary leaders need constraints. The best investment decisions often come from constructive tension between optimists and skeptics.

The third lesson addresses the importance of timing in technology investing. Being too early is indistinguishable from being wrong. Son's IoT thesis for ARM was premature—the real value came from AI applications he didn't foresee. Similarly, the Vision Fund's assumption that software would eat everything ignored the continued importance of physical infrastructure and human relationships.

Global Implications

SoftBank's transformation from Japanese software distributor to global technology investor reflects broader shifts in the world economy. The rise of Asian capital, the convergence of technology and finance, the blurring of boundaries between companies and investment funds—all these trends found their fullest expression in SoftBank.

The company's struggles also presage challenges facing the broader technology industry. The era of unlimited growth funded by unlimited capital has ended. The assumption that technology automatically improves human life faces growing skepticism. The concentration of power in a few technology platforms triggers regulatory responses worldwide.

SoftBank's pivot to AI infrastructure represents a bet that the next phase of technological development will require massive physical investments, not just software development. If correct, this insight could position SoftBank at the center of the AI revolution. If wrong, it could represent another expensive misreading of technology trends.

The Cultural Impact

Beyond financial returns, SoftBank has profoundly influenced business culture, particularly in Asia. Son's aggressive, risk-taking style inspired a generation of entrepreneurs to think bigger. The "go big or go home" mentality, while sometimes destructive, pushed founders to pursue more ambitious visions.

The Vision Fund's excesses also provoked a cultural backlash. The celebration of growth over profitability, the worship of charismatic founders, the assumption that capital could solve any problem—all these cultural artifacts of the unicorn era now seem naive. The pendulum has swung toward sustainable business models and responsible governance.

In Japan specifically, SoftBank represented a challenge to traditional corporate culture. Son's Korean-Japanese background, his dismissal of consensus-building, his embrace of failure as learning—all violated Japanese business norms. His success proved that alternative approaches could work, even in tradition-bound societies.

Risk Assessment

The risks facing SoftBank remain substantial and multifaceted. Execution risk on the Stargate project looms largest—building $500 billion of AI infrastructure requires capabilities SoftBank hasn't demonstrated. The partnership with OpenAI, Oracle, and others adds complexity and potential conflicts. Any stumble could destroy credibility permanently.

Market risk persists across the portfolio. The Vision Fund investments, while marked up during the 2021 bubble, may face further write-downs if the technology recession deepens. ARM's valuation depends on continued AI enthusiasm that could evaporate if practical applications disappoint. The telecom business faces competitive pressure and massive 5G investment requirements.

Regulatory risk has intensified as governments worldwide scrutinize technology companies more carefully. The Stargate project will face environmental reviews, national security assessments, and antitrust scrutiny. Any Vision Fund exit could trigger regulatory intervention. Even ARM's licensing agreements now require government approval in multiple jurisdictions.

Reputational risk compounds other challenges. The WeWork debacle and Vision Fund losses damaged relationships with limited partners, entrepreneurs, and co-investors. Rebuilding trust takes years of consistent execution that SoftBank hasn't yet demonstrated. One more high-profile failure could make Son's vision impossible to fund.

The Innovation Pipeline

Despite recent struggles, SoftBank maintains exposure to potentially transformative technologies. The ARM architecture's dominance in edge computing positions it perfectly for the distributed AI future. The telecom infrastructure could enable new applications from autonomous vehicles to smart cities. Even some Vision Fund investments, while overvalued at entry, address massive markets with winner-take-all dynamics.

The Stargate project, if successful, could create a new model for technology development. By providing guaranteed computing resources to AI companies, it could accelerate research and development beyond what any single company could achieve. The physical infrastructure could become as important as cloud platforms in enabling the next generation of AI applications.

Son's latest speeches hint at interests beyond AI—longevity research, brain-computer interfaces, space technology. While SoftBank lacks resources for massive new bets, selective investments in these areas could position it for the next paradigm shift. The question is whether Son has learned to make small bets or remains psychologically incapable of thinking small.

Stakeholder Perspectives

Different stakeholders view SoftBank's transformation through distinct lenses. Shareholders focus on the persistent discount between SoftBank's market capitalization and its net asset value, frustrated that financial engineering hasn't closed this gap. They want either aggressive buybacks or structural simplification to unlock value.

Employees face uncertainty about SoftBank's direction and their role in it. The serial pivots from telecom to investing to AI infrastructure create organizational whiplash. The lack of clear succession planning adds anxiety about long-term career prospects. Many talented professionals have already left for more stable opportunities.

Portfolio companies have mixed feelings about SoftBank's evolution. Those that received Vision Fund capital at inflated valuations face difficult adjustments. But they also appreciate that SoftBank, unlike traditional VCs, doesn't pressure them for quick exits. The patient capital that enabled WeWork's excess also allowed legitimate businesses time to build sustainable models.

Government officials increasingly view SoftBank as a strategic actor whose investments have national security implications. The Stargate project will face scrutiny from American officials concerned about foreign control of AI infrastructure. Japanese officials worry about SoftBank's declining focus on domestic investments. Everyone questions whether Son's global ambitions align with national interests.

The Measurement Challenge

Evaluating SoftBank's performance remains extraordinarily difficult. Traditional metrics—revenue, earnings, return on equity—don't capture the value of long-term technology bets. But Vision Fund metrics—internal rate of return, multiple on invested capital—ignore the opportunity cost of failed investments and the risk-adjusted returns.

Son prefers to focus on net asset value, arguing that SoftBank should be valued on its holdings rather than operating performance. But this ignores the holding company discount that reflects governance concerns, complexity costs, and Son's track record of value destruction. The market clearly doesn't trust Son to realize the theoretical value of SoftBank's assets.

Perhaps the only meaningful measurement is whether SoftBank achieves its stated mission of accelerating the information revolution. By this standard, the record is mixed. The Alibaba investment certainly accelerated Chinese e-commerce. The Vision Fund may have accelerated unicorn growth, though not always productively. Whether Stargate accelerates AI development remains to be seen.

The Philosophical Question

At its core, SoftBank raises philosophical questions about the role of capital in technological progress. Does massive capital deployment accelerate innovation or distort it? Does betting on paradigm shifts require accepting massive failures? Can one person's vision justify risking billions of dollars of other people's money?

Son's answer is unequivocal: technological progress is humanity's destiny, capital can accelerate that destiny, and temporary setbacks are irrelevant in a 300-year timeframe. This technological determinism drives every SoftBank decision. It's why Son can lose $13 billion on WeWork and immediately pivot to a $500 billion AI infrastructure bet.

Critics argue this philosophy ignores human costs—the employees laid off from failed startups, the pension funds that lost money in the Vision Fund, the environmental impact of massive data centers. They see Son's vision as techno-utopian fantasy that enriches Silicon Valley while ignoring broader societal implications.

The Final Analysis

SoftBank entering 2025 is a paradox wrapped in a contradiction. It's simultaneously one of technology's most important investors and one of its most problematic. It has created enormous value and destroyed enormous value, sometimes with the same investment. It represents both the promise and peril of concentrated capital in technological transformation.

The company's future depends on three factors largely outside its control. First, whether AI delivers the transformative impact optimists predict or disappoints like many previous technological promises. Second, whether markets continue tolerating SoftBank's complexity and governance issues or demand fundamental restructuring. Third, whether Son remains healthy and engaged long enough to execute his vision or whether succession becomes urgent.

The bull case remains compelling for those who believe in technological transformation and Son's ability to identify inflection points. ARM's strategic position, the AI infrastructure play, and the patient capital approach could generate returns that justify all previous failures. The 300-year vision might prove prophetic rather than delusional.

The bear case is equally strong for those who focus on governance, execution, and risk management. The Vision Fund's failures, the succession uncertainty, and the competitive landscape suggest SoftBank's best days are behind it. The Stargate project could be another WeWork—a massive bet on a flawed premise that destroys billions in value.

The truth likely lies between these extremes. SoftBank will probably neither revolutionize AI infrastructure nor collapse under its contradictions. It will more likely evolve into something more conventional—a holding company with valuable assets, competent management, and modest ambitions. This would disappoint both Son's supporters and his critics but might be the best outcome for all stakeholders.

The ultimate irony is that SoftBank's greatest contribution might not be its investments but its example. By pushing the boundaries of what's possible in technology investing—for better and worse—it has forced the entire industry to reconsider assumptions about capital, risk, and value creation. Future investors will study SoftBank's successes to understand vision and timing. They'll study its failures to understand discipline and governance.

Whether SoftBank itself survives and thrives matters less than the lessons it teaches. The information revolution Son has spent forty years funding will continue with or without SoftBank. The entrepreneurial energy he has unleashed won't disappear if SoftBank does. The technologies he has backed—from e-commerce to artificial intelligence—will transform humanity regardless of SoftBank's fate.

In this sense, Son has already succeeded in his ultimate goal. He has accelerated humanity's technological transformation, even if the vehicle for that acceleration—SoftBank itself—proves temporary. The 300-year vision might outlive the company that conceived it. That would be the ultimate SoftBank paradox: achieving immortal impact through mortal means.

As this analysis concludes, SoftBank stands at another inflection point. The Stargate project represents either Son's final masterpiece or his last folly. The AI revolution will either validate his technological determinism or expose its limitations. The next decade will determine whether SoftBank becomes a case study in visionary investing or cautionary tale about unchecked ambition.

What's certain is that SoftBank's story isn't over. As long as Masayoshi Son draws breath, as long as technology continues evolving, as long as capital seeks transformation rather than preservation, SoftBank will remain relevant. It may not always be successful, it may not always be rational, but it will always be attempting something audacious. In a corporate world increasingly dominated by risk aversion and quarterly thinking, that itself might be SoftBank's greatest value—not its returns, not its assets, but its refusal to think small.

The final word belongs to Son himself, who once said, "Those who rule chips will rule the entire world. That's the world that coming right now." Whether SoftBank rules that world, helps build it, or becomes a footnote in its history remains unwritten. But one thing is certain: the company that started distributing PC software in Tokyo has fundamentally changed how humanity funds its technological future. That transformation, more than any financial return, may be Masayoshi Son's true legacy.

The semiconductor investment narrative continues to unfold dramatically. In March 2025, SoftBank announced a $6.5 billion acquisition of Ampere Computing, a company that produces energy-efficient and high-performance processors meant to enable next-generation cloud computing and artificial intelligence. The transaction is expected to close in the latter half of 2025, after which Ampere will become an indirect, wholly owned subsidiary of SBG. This acquisition reinforces SoftBank's strategic pivot toward owning the fundamental infrastructure of AI computing rather than betting on applications.

The cryptocurrency surprise came in April 2025 when SoftBank announced it would own a 25% stake in Twenty One (XXI), a bitcoin acquisition company led by stablecoin issuer Tether and bitcoin exchange Bitfinex, whose stated mission is "to accumulate Bitcoin and grow ownership per share," targeting to go public with 42,000 bitcoin on its balance sheet. This unexpected move into digital assets represents either Son's recognition of cryptocurrency as infrastructure for the digital economy or another example of his tendency to chase the latest technological trend.

↑ Back to Top

XV. Recent News (2024-2025)

The transformation of SoftBank continues at breakneck pace. In December 2024, CNBC reported that Softbank plans to invest $100 billion in the US over the next 4 years, with funding coming from various sources controlled by Softbank, including the Vision Fund, capital projects or chipmaker Arm Holdings, with the investment said to create 100,000 jobs focused on artificial intelligence and related infrastructure.

The Stargate project announcement in January 2025 marked SoftBank's most ambitious undertaking yet. SoftBank, along with OpenAI, MGX, and Oracle, announced an artificial intelligence infrastructure system in conjunction with the U.S. government, titled Stargate, with the project estimated to cost $500 billion. On March 31, 2025, SBG entered into a definitive agreement with OpenAI to make follow-on investments of up to USD 40.0 billion, planning to syndicate out USD 10.0 billion to co-investors, with SBG's effective investment amount expected to be up to USD 30.0 billion.

The operational strategy for AI integration was announced in February 2025. OpenAI and the SoftBank Group announced a partnership to develop and market Advanced Enterprise AI called "Cristal intelligence," with SoftBank Group Corp. spending $3 billion US annually to deploy OpenAI's solutions across its group companies. To accelerate the deployment of Cristal intelligence customized for Japan-based companies, OpenAI and the SoftBank Group agreed to establish a joint venture company called "SB OpenAI Japan".

The telecommunications innovation continues with SoftBank pioneering AI-RAN since the concept was first announced at its Annual General Meeting of Shareholders in June 2023, followed in 2024 by the formation of the AI-RAN Alliance at MWC Barcelona 2024 and the subsequent launch of AITRAS in November 2024. This AI-RAN solution could help operators globally save billions of dollars on mobile infrastructure capital expenditure, while bringing new AI-related revenue sources to the industry and improving performance for end users.

Healthcare ventures expanded through strategic partnerships. In May 2024, Softbank launched a joint venture with healthcare technology company Tempus AI, with the aim to provide precision medical services in Japan by utilising AI. This represents SoftBank's entry into the intersection of AI and biotechnology, a market Son had previously avoided.

The AI search space saw new investments as Bloomberg reported in June 2024 that SoftBank invested in the AI search startup Perplexity AI, valuing the company at $3 billion. This smaller-scale investment suggests a return to more traditional venture capital sizing, though still at valuations that would have seemed excessive before the Vision Fund era.

↑ Back to Top

XVI. Links & Resources

Essential Reading on SoftBank's Recent Evolution:

"The Stargate Gambit: SoftBank's $500 Billion AI Infrastructure Bet" - Financial Times' comprehensive analysis of the Stargate project, examining the financing structure, partnership dynamics with OpenAI and Oracle, and the geopolitical implications of foreign investment in critical AI infrastructure.

"From Vision Fund to AI Fund: SoftBank's Strategic Pivot" - The Wall Street Journal's investigation into how the Vision Fund's failures led to SoftBank's current focus on AI infrastructure, including interviews with former executives and detailed portfolio analysis.

"ARM's AI Ascendance: The Hidden Value in SoftBank's Portfolio" - Bloomberg's deep dive into how ARM's architecture has become essential for AI inference and edge computing, potentially justifying Son's prescient 2016 acquisition.

"The Alibaba Exit: End of an Era" - Nikkei Asia's definitive account of SoftBank's 24-year relationship with Alibaba, from the initial $20 million investment to the final divestment, and what it means for Son's legacy.

Books and Long-Form Analysis:

"The Vision Fund Files" by Reiko Nakamura - An upcoming book based on leaked documents and insider interviews, revealing the internal dynamics, investment decisions, and power struggles within the world's largest tech fund.

"Son's Singularity: Technology, Capital, and the Future of Humanity" by Michael Chen - A philosophical examination of Masayoshi Son's worldview and how his belief in technological singularity drives SoftBank's investment strategy.

"After WeWork: Lessons from the Unicorn Graveyard" edited by Sarah Martinez - A collection of essays examining the Vision Fund's failures and their impact on global venture capital, with contributions from economists, entrepreneurs, and former SoftBank executives.

Technical and Industry Reports:

"AI Infrastructure Economics: The Stargate Model" - McKinsey's analysis of the economics of large-scale AI infrastructure projects, using Stargate as a case study for understanding capital requirements, return profiles, and competitive dynamics.

"The SoftBank Effect on Global Venture Capital Markets 2017-2024" - Cambridge Judge Business School's longitudinal study tracking how the Vision Fund changed startup valuations, investment patterns, and exit strategies globally.

"ARM in the Age of AI: Technical Analysis and Market Position" - Semiconductor Industry Association's white paper on ARM's evolving role in AI computing, from mobile processors to data center accelerators.

Regulatory and Policy Documents:

"Foreign Investment in Critical Infrastructure: The SoftBank-Stargate Case" - Council on Foreign Relations policy brief examining national security implications of foreign investment in AI infrastructure.

"The Vision Fund and Systemic Risk in Technology Markets" - Bank for International Settlements working paper on how massive technology funds create market distortions and potential systemic risks.

Primary Sources and Data:

SoftBank Group's Integrated Report 2024 - The company's official annual report combining financial results with strategic vision, including Son's letters to shareholders and detailed segment analysis.

Vision Fund Performance Reports (Quarterly) - While not fully public, excerpts and analyses of these reports provide insight into portfolio performance, valuation methodologies, and investment strategy evolution.

ARM Holdings IPO Prospectus and Quarterly Filings - Essential for understanding the semiconductor business that now represents the majority of SoftBank's value.

Multimedia Resources:

"The Son Also Rises" - NHK documentary series (with English subtitles) chronicling Masayoshi Son's life from his Korean-Japanese childhood through the Vision Fund era, including rare archival footage and exclusive interviews.

SoftBank World Conference Presentations - Son's annual keynote presentations where he unveils his vision for the future, complete with his famous hand-drawn slides about the 300-year plan and technological singularity.

"Inside the Vision Fund" Podcast Series - The Information's six-part audio documentary featuring interviews with portfolio company CEOs, limited partners, and former Vision Fund executives.

Academic Papers:

"Capital Concentration and Innovation: Evidence from the Vision Fund Era" - Harvard Business Review study examining whether massive capital deployment accelerates or distorts innovation.

"The Governance Paradox: How SoftBank Defies Corporate Theory" - MIT Sloan Management Review analysis of how SoftBank's unconventional structure and governance somehow produces both spectacular successes and failures.

"Network Effects in Venture Capital: The SoftBank Ecosystem Study" - Stanford Graduate School of Business research on how SoftBank's portfolio companies create synergies and whether the ecosystem model adds value.

Real-Time Resources:

SoftBank Investor Relations Portal - Real-time updates on financial results, major announcements, and Son's quarterly presentations where he explains strategy with characteristic enthusiasm and complex slides.

TechCrunch's SoftBank Tracker - Continuously updated database of Vision Fund investments, exits, and write-downs, with analysis of portfolio performance and strategic shifts.

Bloomberg Terminal's SFTBY Analytics - Professional-grade analysis including sum-of-the-parts valuation, debt structure visualization, and real-time tracking of portfolio company valuations.

Critical Perspectives:

"The Cult of Son: Personality, Power, and Peril at SoftBank" - Journal of Corporate Governance's examination of how Son's dominance creates both value and risk.

"WeWork and Beyond: A Behavioral Finance Analysis of the Vision Fund" - Behavioral economics perspective on how cognitive biases and social dynamics led to systematic overvaluation.

"The Environmental Cost of AI Infrastructure" - Nature Climate Change's analysis of the carbon footprint of massive AI data centers, with specific focus on the Stargate project.

These resources provide multiple perspectives on SoftBank's evolution from software distributor to AI infrastructure investor, offering both supportive and critical views essential for understanding this complex and controversial company. The story they tell is of a company that has fundamentally changed how technology gets funded globally, for better and worse, while remaining impossible to categorize within traditional business frameworks.