SoftBank Group has secured a $40 billion unsecured bridge financing facility – the largest AI-focused corporate loan in history – to fund a $30 billion follow-on investment in OpenAI and accelerate its dominance over the artificial intelligence infrastructure race. Announced on March 27, 2026, the deal was syndicated by JPMorgan Chase, Goldman Sachs, Mizuho Bank, Sumitomo Mitsui Banking Corporation, and MUFG Bank, and carries a 12-month maturity date of March 25, 2027. The loan’s unsecured nature and short-term structure signal both extraordinary lender confidence in SoftBank’s AI thesis and widespread market expectation that an OpenAI IPO could arrive before year-end.

The financing pushes SoftBank’s cumulative investment in OpenAI to $64.6 billion, representing approximately 13% ownership in the company that has become the standard-bearer for generative AI. It follows SoftBank’s initial $30 billion commitment as part of OpenAI’s record-shattering $110 billion funding round concluded in February 2026. With this deal, Masayoshi Son’s conglomerate has placed the single largest private bet on artificial intelligence in corporate history, dwarfing even the Vision Fund’s original $100 billion deployment a decade ago.

Inside the $40 Billion Loan: Structure, Terms, and Why Banks Said Yes

The sheer scale of SoftBank’s $40 billion bridge facility demands scrutiny. Unsecured loans of this magnitude are exceedingly rare in corporate finance, and the fact that five of the world’s largest banks agreed to syndicate the deal without collateral speaks volumes about how the financial establishment views the AI opportunity – and SoftBank’s position within it.

JPMorgan Chase and Goldman Sachs led the syndication on the Western side, while three of Japan’s largest banks – Mizuho, SMBC, and MUFG – rounded out the consortium. The 12-month maturity window is unusually short for a facility of this size, suggesting that both SoftBank and its lenders anticipate a liquidity event, most likely an OpenAI IPO, within that timeframe.

“This is not a conventional corporate loan – it’s essentially a bridge to an IPO,” said Dan Ives, Managing Director at Wedbush Securities. “The banks are betting that OpenAI goes public within 12 months at a valuation that could exceed $400 billion, which would make SoftBank’s 13% stake worth more than $50 billion. The math works.”

The unsecured structure is particularly notable given SoftBank’s financial history. The company carried approximately $130 billion in consolidated debt as of Q3 FY2025, though its asset base – anchored by its ARM Holdings subsidiary and growing AI portfolio – provides substantial implicit backing. SoftBank’s shares rose 3.24% on the Tokyo Stock Exchange following the announcement, adding roughly $6.5 billion in market capitalization in a single session.

For context, the largest unsecured corporate loan facilities in tech history include Microsoft’s $60 billion revolving credit facility (2021), Apple’s $10 billion commercial paper program, and Broadcom’s $40 billion acquisition financing for VMware in 2023. SoftBank’s deal matches Broadcom’s in raw size but carries significantly more risk given its concentration in a single, pre-IPO company.

SoftBank’s $64.6 Billion OpenAI Position: The Biggest Private Bet in AI History

With the follow-on investment, SoftBank will have poured a total of $64.6 billion into OpenAI – a sum that exceeds the GDP of more than 100 countries. The investment was made through SoftBank’s Vision Fund 2 vehicle and builds on the $30 billion initial commitment that anchored OpenAI’s $110 billion mega-round in February 2026.

That funding round valued OpenAI at approximately $300 billion, making it the most valuable private company in history. The additional $30 billion follow-on suggests an even higher implied valuation, potentially in the $350-400 billion range, though exact terms have not been publicly disclosed.

“Masayoshi Son is doing what he’s always done – making a concentrated, conviction-driven bet on what he believes is the defining technology of the next century,” said Aswath Damodaran, Professor of Finance at NYU Stern School of Business. “The difference this time is that the magnitude is unprecedented. A $64.6 billion position in a single company is extraordinary even by SoftBank standards.”

SoftBank’s 13% ownership stake positions it as one of OpenAI’s largest shareholders, behind only Microsoft, which holds a complex revenue-sharing arrangement and estimated 49% economic interest. The investment gives SoftBank significant influence over OpenAI’s strategic direction, including its transition from a capped-profit structure to a full for-profit corporation – a process that is expected to complete before any IPO.

The Stargate Connection: SoftBank’s $500 Billion AI Infrastructure Vision

The $40 billion loan cannot be understood in isolation. It represents the financial backbone of the Stargate Project, the $500 billion AI infrastructure initiative announced by President Donald Trump on January 21, 2025, at a White House press conference. Stargate LLC is a joint venture led by SoftBank and OpenAI, with Oracle and MGX as initial equity funders, and ARM, Microsoft, and NVIDIA as key technology partners.

Under the Stargate structure, SoftBank handles financial responsibility while OpenAI manages operational execution. Masayoshi Son serves as chairman of the venture. The project has committed to an initial $100 billion deployment, with plans to invest a total of $500 billion by 2029 in U.S.-based AI data centers and infrastructure.

Construction is already underway. As of March 2026, 10 data centers are under construction in Abilene, Texas, with 4.5 GW of power capacity added recently, bringing the total toward a 10 GW goal. The project is expected to create over 100,000 jobs and is positioned as a cornerstone of the administration’s strategy to maintain American leadership in AI.

“Stargate is the most ambitious infrastructure project in the history of the technology industry,” said Patrick Moorhead, CEO and Chief Analyst at Moor Insights & Strategy. “To put 10 GW in perspective, that’s roughly equivalent to the power output of 10 nuclear reactors. The scale is unlike anything we’ve seen in Silicon Valley’s history.”

The AI data center power crisis makes Stargate’s energy ambitions particularly significant. With Big Tech collectively requiring an estimated 125 GW of data center capacity by the end of the decade, Stargate’s 10 GW target represents a substantial portion of the industry’s total needs. The project’s Texas location was chosen in part for access to relatively affordable wind and natural gas power, as well as favorable state regulations.

How SoftBank’s AI Gamble Compares to Big Tech’s Spending Spree

SoftBank’s $64.6 billion OpenAI investment must be viewed against the backdrop of Big Tech’s $700 billion AI infrastructure spending race in 2026. Every major technology company is pouring capital into AI at historic rates, but SoftBank’s approach is fundamentally different: rather than building its own AI capabilities, it is betting on being the financial kingmaker behind the company it believes will win.

Company	2026 AI Investment	Strategy	Primary Focus	Risk Level
SoftBank (OpenAI)	$64.6B cumulative	Concentrated equity bet	OpenAI ownership + Stargate	Very High
Microsoft	$80B capex planned	Infrastructure + integration	Azure AI + Copilot ecosystem	Medium
Alphabet/Google	$75B capex planned	Vertical integration	Gemini models + Cloud AI	Medium
Meta	$60-65B capex planned	Infrastructure + open source	Llama models + AI products	Medium-High
Amazon/AWS	$100B+ capex planned	Cloud infrastructure	AWS AI services + custom chips	Medium
Apple	$5B estimated	On-device AI	Apple Intelligence	Low

The contrast is stark. While Microsoft, Google, Amazon, and Meta are diversifying their AI investments across infrastructure, models, and applications, SoftBank has placed an enormous concentrated bet on a single company. This is consistent with Masayoshi Son’s investment philosophy – he has long argued that identifying and backing the dominant player in a technological revolution generates returns that diversified approaches cannot match.

Microsoft’s $150 billion AI capex commitment is spread across Azure infrastructure, custom chips, and its Copilot product ecosystem. Google and Meta are building both models and infrastructure in-house. Amazon is investing heavily in custom Trainium and Graviton chips alongside expanded AWS AI services. SoftBank, by contrast, is essentially financing the entire AI revolution through a single equity position.

SoftBank’s Vision Fund Track Record: Lessons from Boom and Bust

SoftBank’s massive AI bet carries particular weight given the company’s volatile investment history. The original Vision Fund, launched in 2017 with over $100 billion in capital – including $45 billion from Saudi Arabia’s Public Investment Fund – was the world’s largest technology investment vehicle. Its performance has been a roller coaster that offers both cautionary tales and validation for Son’s approach.

The Vision Fund reached peak valuations of $154 billion across both Fund 1 and Fund 2 in March 2021, recording a $36.99 billion annual profit driven by investments like Coupang and DoorDash. But the tide turned dramatically: the fund posted a record ¥3.5 trillion ($27.4 billion) loss in the fiscal year ending March 2022, followed by an even larger $32 billion loss the following year. By early 2023, quarterly investment activity had plummeted 90% year-over-year, falling to just $300 million.

The WeWork debacle remains the most prominent cautionary tale. SoftBank invested approximately $18.5 billion in the coworking company, which peaked at a $47 billion valuation before its spectacular implosion. Son later described the investment as “foolish,” but the experience has not dampened his appetite for concentrated bets.

More recently, however, the Vision Funds have staged a significant recovery. In Q1 FY2025 (April-June 2025), SoftBank Vision Funds recorded a gain of ¥660.2 billion, a ¥627.8 billion improvement year-over-year. Vision Fund 1 contributed ¥517.4 billion in gains while Vision Fund 2 added ¥114.5 billion, driven primarily by AI-related portfolio appreciation.

“Son’s track record is deeply polarizing, but there’s a pattern that critics often miss,” said Cathie Wood, CEO of ARK Invest. “He was early and right on the mobile internet with his Vodafone Japan and Yahoo Japan bets. He was early and wrong with WeWork. The question with OpenAI is whether the AI revolution is closer to the mobile internet – a genuine platform shift – or closer to WeWork – a story that outran its economics. We believe it’s the former.”

The ARM Factor: SoftBank’s $170 Billion AI Use Point

Underpinning SoftBank’s entire AI strategy is its majority ownership of ARM Holdings, the chip architecture company whose designs power virtually every smartphone and an increasing share of data center processors. SoftBank acquired ARM for $32 billion in 2016 and took it public in September 2023 at a $54.5 billion valuation. By early 2026, ARM’s market capitalization has grown to approximately $170 billion, making it one of the most successful technology investments in history.

SoftBank retains approximately 90% ownership of ARM, giving it effective control over the company and a stake worth roughly $153 billion at current valuations. This position serves as implicit collateral for SoftBank’s borrowing activities, even when loans are technically unsecured. ARM’s importance to the AI ecosystem – its 136-core AGI CPU marks a historic shift into AI data center silicon – means that SoftBank’s two largest bets are deeply interconnected.

ARM’s architecture is a key technology partner in the Stargate Project, and the company’s designs are expected to power a significant portion of the data center infrastructure that Stargate will build. This creates a virtuous cycle: SoftBank funds Stargate, which builds data centers that use ARM processors, which drives ARM’s revenue and stock price, which supports SoftBank’s balance sheet and borrowing capacity.

OpenAI’s Path to IPO: What the $40 Billion Loan Reveals

The 12-month maturity of SoftBank’s bridge loan is the strongest signal yet that OpenAI’s IPO is imminent. If SoftBank needs to repay $40 billion by March 2027, it requires either an IPO that unlocks liquidity, a secondary share sale, or a refinancing – and the market overwhelmingly expects option one.

OpenAI has been steadily preparing for a public listing. The company’s transition from its original capped-profit structure to a full for-profit corporation is expected to complete in 2026, removing one of the last structural obstacles to an IPO. Revenue has been growing rapidly, with estimates placing OpenAI’s annualized revenue run rate above $12 billion as of early 2026, up from $3.4 billion in late 2024.

An IPO at a $400 billion valuation – the figure many analysts now consider the base case – would value SoftBank’s 13% stake at approximately $52 billion, providing ample liquidity to repay the $40 billion loan and generating a significant return on its cumulative $64.6 billion investment when combined with the appreciated equity.

However, the stakes are enormous. If an IPO is delayed, market conditions deteriorate, or OpenAI’s competitive position weakens, SoftBank could face a liquidity crunch. The $40 billion loan is unsecured, but lenders would still expect repayment, potentially forcing SoftBank to sell other assets – including portions of its ARM stake – at unfavorable prices.

“The risk-reward here is binary,” said Gil Luria, Head of Technology Research at D.A. Davidson. “If OpenAI goes public at the valuations the market expects, this will be remembered as one of the greatest financial maneuvers in corporate history. If something goes wrong – regulatory delays, competitive disruption, a market downturn – it could threaten SoftBank’s entire corporate structure.”

The Competitive Landscape: Why SoftBank Is Doubling Down Now

SoftBank’s urgency is partly explained by the intensifying competition in the AI space. OpenAI’s lead, once considered insurmountable, is being challenged from multiple directions. Google’s Gemini platform has reached 750 million users as of March 2026. Anthropic, valued at approximately $80 billion, continues to push the frontier with its Claude models. And DeepSeek’s open-source models have demonstrated that state-of-the-art performance can be achieved at a fraction of the cost.

OpenAI has responded by expanding aggressively across multiple fronts. Its Pentagon deal opened the defense market. Its GPT-5.4 Mini and Nano subagent models are targeting the enterprise automation market. And its Titan custom chip program with Samsung aims to reduce dependence on NVIDIA and lower inference costs.

SoftBank’s additional capital injection provides OpenAI with the resources to compete on all these fronts simultaneously. The $30 billion follow-on investment, combined with Stargate’s infrastructure buildout, is designed to create a vertically integrated AI powerhouse that controls everything from chip design to model training to end-user applications.

AI Company	Latest Valuation	Key Investor	2026 Revenue Est.	Key Differentiator
OpenAI	~$300-400B	SoftBank ($64.6B)	$12B+ ARR	GPT models + Stargate infra
Anthropic	~$80B	Google, Amazon	$4B+ ARR	Constitutional AI + safety focus
Google DeepMind	Part of Alphabet ($2.2T)	Alphabet (internal)	Integrated into Cloud	Gemini + Search integration
xAI (Grok)	~$50B	Elon Musk + investors	$1B+ estimated	X platform integration
DeepSeek	Not publicly valued	High-Flyer Capital	Not disclosed	Cost-efficient open source
Meta AI (Llama)	Part of Meta ($1.7T)	Meta (internal)	Integrated into apps	Open source + social integration

Market Reaction and Analyst Sentiment

The market’s response to SoftBank’s $40 billion loan has been cautiously optimistic. SoftBank shares gained 3.24% on the Tokyo Stock Exchange on March 27, reflecting investor confidence that the deal strengthens rather than threatens the company’s financial position. The broader AI sector also received a boost, with the Philadelphia Semiconductor Index rising 1.8% and major AI stocks including NVIDIA, ARM, and Oracle all posting gains.

However, not all analysts are bullish. The concentration risk inherent in SoftBank’s strategy – with nearly $65 billion tied to a single pre-IPO company – has drawn comparisons to the Vision Fund’s earlier concentrated bets on companies like WeWork and Uber. Credit rating agencies are monitoring the situation closely, with Moody’s noting that SoftBank’s use ratios could come under pressure if AI market conditions shift.

The five-bank syndicate structure itself provides some reassurance. JPMorgan Chase, the world’s largest bank by assets, would not have led the facility without extensive due diligence on both SoftBank’s creditworthiness and OpenAI’s IPO prospects. Goldman Sachs’ involvement suggests the investment bank may be positioning for a lead role in OpenAI’s eventual public offering.

Bond markets have been more measured. SoftBank’s existing corporate bonds traded slightly lower following the announcement, reflecting some concern about additional debt loading on the balance sheet. The company’s credit default swap spreads widened by approximately 15 basis points, suggesting a modest increase in perceived credit risk.

The Geopolitical Dimension: AI Sovereignty and National Competition

SoftBank’s investment carries significant geopolitical implications. The Stargate Project’s explicit goal of maintaining American AI leadership positions SoftBank – a Japanese conglomerate – as a critical intermediary between the U.S. government’s AI ambitions and the capital needed to realize them. The White House’s endorsement of the project, with Trump personally announcing it at a press conference, underscores the degree to which AI infrastructure has become a matter of national security.

China’s AI ambitions provide the competitive backdrop. Despite U.S. export controls on advanced chips, Chinese companies like DeepSeek and Baidu continue to make significant progress in AI model development. ByteDance’s $2.5 billion procurement of NVIDIA B200 chips through Malaysia demonstrates that Chinese firms are finding creative workarounds to access cutting-edge hardware.

SoftBank’s position is uniquely complex. As a Japanese company investing in American AI infrastructure while maintaining deep relationships across Asia, it serves as a bridge between geopolitical blocs in ways that purely American or Chinese companies cannot. Masayoshi Son’s personal relationships with both U.S. and Asian political leaders give SoftBank a diplomatic dimension that purely financial analysis misses.

The European Union, meanwhile, has been notably absent from the mega-scale AI infrastructure investments that are defining the 2026 landscape. The EU’s AI Act regulatory framework has been criticized for focusing on risk mitigation rather than capability building, and no European company or consortium has announced anything approaching the scale of Stargate.

Risks and Potential Downsides of SoftBank’s Strategy

For all its potential upside, SoftBank’s $40 billion loan and $64.6 billion OpenAI position carry substantial risks that investors and market observers should carefully consider.

Concentration risk is the most obvious concern. Having nearly $65 billion exposed to a single pre-IPO company means that any setback at OpenAI – whether competitive, regulatory, or operational – would have an outsized impact on SoftBank’s financial health. The Vision Fund’s history demonstrates how quickly sentiment can shift: the portfolio went from $154 billion in fair value to recording $32 billion annual losses in just two years.

IPO timing risk is equally significant. The 12-month loan maturity creates a hard deadline. If regulatory reviews, market conditions, or OpenAI’s corporate restructuring delay the IPO beyond March 2027, SoftBank would need to refinance the facility – potentially on less favorable terms – or liquidate other assets. Given that SoftBank’s most liquid major asset is its ARM stake, the company could be forced to sell ARM shares at a time not of its choosing.

AI market saturation represents a longer-term concern. The $700 billion-plus that Big Tech companies have collectively committed to AI infrastructure in 2026 assumes sustained demand growth for AI compute. If enterprise AI adoption plateaus, or if efficiency improvements reduce the need for brute-force compute scaling, the infrastructure that Stargate and others are building could become overcapacity.

Regulatory risk looms over the entire AI sector. Antitrust scrutiny of OpenAI’s relationships with Microsoft and now SoftBank could slow or complicate the IPO process. The FTC has already indicated interest in examining whether large AI investments constitute anti-competitive arrangements, and any enforcement action could introduce delays that SoftBank’s 12-month loan timeline cannot accommodate.

Five Predictions for the SoftBank-OpenAI Partnership

Based on the trajectory of SoftBank’s investment and the broader AI market dynamics, here are five predictions for how this story unfolds over the next 12-18 months:

1. OpenAI will IPO in Q4 2026 or Q1 2027. The 12-month loan maturity all but guarantees that OpenAI will pursue a public listing before March 2027. Expect the S-1 filing by mid-2026, with a targeted listing in the October-December window when market conditions are typically favorable for large tech IPOs. The initial valuation target will be $350-450 billion.

2. SoftBank will become OpenAI’s largest shareholder post-IPO. As OpenAI restructures from its capped-profit model to a full for-profit corporation, the equity dynamics will shift. Microsoft’s revenue-sharing arrangement may convert to a more conventional equity stake, but SoftBank’s 13% ownership – acquired through direct equity investment – positions it to emerge as the single largest shareholder by voting power.

3. Stargate will announce expansion beyond Texas by year-end 2026. With 10 data centers under construction in Abilene, the project will likely announce additional sites in states with favorable energy profiles and regulatory environments. Arizona, Virginia, and Ohio are the most probable candidates, given their existing data center ecosystems and power availability.

4. SoftBank will face a credit rating review within six months. The addition of $40 billion in unsecured debt, on top of existing use, will likely trigger a review from major rating agencies. While a downgrade is not guaranteed – the ARM stake provides substantial asset coverage – the agencies will want assurance that the IPO timeline is realistic.

5. At least one additional major investor will join the Stargate consortium by mid-2026. The project’s $500 billion ambition requires capital beyond what SoftBank and OpenAI can provide alone. Expect a sovereign wealth fund – most likely from the UAE, Saudi Arabia, or Singapore – to join as a major equity partner, bringing both capital and geopolitical diversification.

What This Means for the Broader AI Industry

SoftBank’s $40 billion loan sends a clear message to the AI industry: the era of cautious, incremental investment is over. The willingness of five major banks to extend $40 billion unsecured to fund an AI bet suggests that the financial establishment views artificial intelligence not as a speculative bubble but as a fundamental economic transformation on par with the internet or electrification.

For OpenAI’s competitors, the deal raises the competitive stakes. Anthropic, despite its strong technological position, operates at a fraction of OpenAI’s resource base. Google and Meta have the financial resources to compete but face the organizational challenges of incumbent companies. Open-source challengers like DeepSeek and Meta’s Llama models offer cost-competitive alternatives but lack the infrastructure moat that Stargate is building for OpenAI.

The semiconductor industry stands to benefit enormously. NVIDIA, ARM, and other chip makers will supply the processors for Stargate’s data centers. NVIDIA’s $4 billion silicon photonics investments in Lumentum and Coherent are designed to solve the data center connectivity challenges that projects like Stargate create. Samsung’s $73 billion semiconductor investment program is partly motivated by the memory chip demand that AI data centers generate.

For enterprise customers, the influx of capital into AI infrastructure should translate to declining compute costs over the medium term. As Stargate and similar projects add capacity, the supply-demand balance for AI compute will shift from the current seller’s market to a more competitive environment, potentially accelerating enterprise AI adoption.

Historical Context: How Son’s AI Bet Compares to His Greatest Gambles

Masayoshi Son has a well-documented history of making enormous, concentrated bets on technological transformations. His $20 million investment in Alibaba in 2000 grew to a $130 billion position at its peak – one of the most profitable venture investments in history. His $22 billion acquisition of Sprint in 2013 was less successful, eventually resulting in a merger with T-Mobile that diluted SoftBank’s position. The Vision Fund’s $100 billion deployment across hundreds of companies produced wildly uneven results, from spectacular wins like Coupang to devastating losses like WeWork.

The OpenAI bet is distinct from all of these in its sheer concentration. With $64.6 billion committed to a single company, Son is not hedging across a portfolio – he is making the single largest investment in a single company in the history of private markets. If OpenAI becomes the defining platform company of the AI era, this will be remembered as the greatest investment call in corporate history. If it doesn’t, the consequences for SoftBank could be existential.

Son has famously described artificial intelligence as the single most important invention in human history, surpassing even the internet. At SoftBank’s annual shareholder meeting in June 2025, he predicted that artificial superintelligence – AI systems that surpass human cognitive capabilities across all domains – would arrive within 10 years and that SoftBank’s purpose was to be the financial architect of that transformation.

The Banking Angle: Why Five Banks Bet $40 Billion on AI

The five-bank syndicate behind SoftBank’s loan represents a cross-section of global finance, and each institution has its own strategic motivations for participating.

JPMorgan Chase, the world’s largest bank by assets with over $4 trillion on its balance sheet, has been aggressively expanding its technology lending practice. The bank views AI infrastructure financing as a multi-decade opportunity and wants to establish itself as the go-to lender for AI mega-projects.

Goldman Sachs is positioning for a dual role: lender today, IPO underwriter tomorrow. By participating in the loan facility, Goldman is strengthening its relationship with both SoftBank and OpenAI ahead of what could be the largest technology IPO in history.

Mizuho, SMBC, and MUFG – Japan’s three largest banking groups – have long-standing relationships with SoftBank and deep familiarity with its financial structure. Their participation reflects both institutional loyalty and a calculated assessment that SoftBank’s AI thesis, backed by its ARM holdings and OpenAI position, represents acceptable risk.

The participation of all three major Japanese banks is particularly significant. It signals that Japan’s financial establishment – historically conservative – has embraced AI as a strategic priority and is willing to extend unprecedented credit to fund it. This alignment between Japan’s tech and financial sectors could have lasting implications for the country’s role in the global AI economy.

Related Coverage

Further Reading on AI Investment and Infrastructure

• OpenAI’s $110 Billion Funding Round: Inside the Largest Private Investment in History
• Big Tech’s $700 Billion AI Infrastructure Bet: Inside the 2026 Spending Race
• Microsoft’s $150 Billion AI Capex Gamble: Inside the Azure Surge
• The AI Data Center Power Crisis: How Big Tech’s 125 GW Appetite Is Reshaping the US Energy Grid
• Arm’s AGI CPU: Inside the 136-Core Chip Marks a Historic Shift to AI Data Center Silicon
• OpenAI’s Pentagon Deal: Inside the Classified Military AI Agreement
• NVIDIA’s $4 Billion Silicon Photonics Bet: Reshaping AI Data Center Connectivity

Frequently Asked Questions

How much did SoftBank borrow and what is the loan for?

SoftBank secured a $40 billion unsecured bridge financing facility on March 27, 2026. The primary purpose is to fund a $30 billion follow-on investment in OpenAI, with the remainder available for general corporate purposes including the Stargate AI infrastructure project. The loan has a 12-month maturity, expiring on March 25, 2027.

How much has SoftBank invested in OpenAI in total?

With the $30 billion follow-on investment, SoftBank’s cumulative investment in OpenAI reaches $64.6 billion. This includes its initial $30 billion commitment as part of OpenAI’s $110 billion funding round and additional investments through Vision Fund 2. SoftBank now holds approximately 13% ownership of OpenAI.

Which banks are backing the $40 billion loan?

Five major global banks syndicated the facility: JPMorgan Chase, Goldman Sachs, Mizuho Bank, Sumitomo Mitsui Banking Corporation (SMBC), and MUFG Bank. The participation of both leading Western and Japanese banks reflects broad institutional confidence in SoftBank’s AI investment thesis.

What is the Stargate Project and how does it relate to this loan?

The Stargate Project is a $500 billion AI infrastructure joint venture between SoftBank and OpenAI, announced by President Trump in January 2025. It aims to build AI data centers across the United States, with 10 facilities already under construction in Abilene, Texas. SoftBank’s $40 billion loan partially funds its financial obligations to the project.

Does this loan signal an OpenAI IPO?

The 12-month loan maturity (March 2027) is widely interpreted as a strong signal that OpenAI plans to go public before that deadline. An IPO would provide SoftBank with the liquidity needed to repay the facility. Analysts estimate that an IPO at a $400 billion valuation would value SoftBank’s 13% stake at approximately $52 billion.

What are the biggest risks of SoftBank’s strategy?

The primary risks include concentration risk ($64.6 billion in a single pre-IPO company), IPO timing risk (the 12-month loan deadline), potential AI market saturation, and regulatory uncertainty. SoftBank’s Vision Fund history – which includes both spectacular gains and record $32 billion annual losses – illustrates the volatility inherent in concentrated technology bets.

How does SoftBank’s investment compare to other Big Tech AI spending?

While Microsoft plans $80 billion, Google $75 billion, and Amazon over $100 billion in 2026 AI capex, SoftBank’s approach is fundamentally different. Big Tech companies are investing in their own infrastructure and products, while SoftBank has placed a concentrated equity bet on OpenAI becoming the dominant AI platform. SoftBank’s $64.6 billion represents the largest single-company AI investment in history.