Is AI infrastructure a bubble or a "group buy for time"? Analyzing the financial structure behind the 3 trillion dollars.

This is not a simple binary debate of “bubble vs non-bubble”; the answer may be more complex and sophisticated than you think. I do not have a crystal ball to foresee the future. But I try to delve into the underlying financial structure of this feast and construct a set of analytical frameworks.

The article is lengthy and has many details, so let's start with the conclusion:

• In terms of direction, I do not think this is a big bubble. However, there are high risks in certain segments.
• To be more precise, the current AI infrastructure resembles a long march of “banding together + buying time”. Major companies (( Microsoft, Google, Meta, Nvidia, etc.) leverage financial engineering to unlock massive leverage, while outsourcing the primary credit risk to special purpose vehicles (SPVs) and the capital markets, tightly binding the interests of all participants.
• The so-called “buying time” refers to betting on whether their cash flow and external resources can hold out until the day when “AI truly enhances productivity” arrives.
• If the bet wins, AI will fulfill its promise, and the big companies will be the biggest winners. If the bet loses (AI's progress is not as expected or costs are too high), the first to be hurt will be the external resources providing financing.
• This is not the kind of “bank leverage excess, single-point detonation” bubble of 2008. This is a massive experiment in direct financing, led by the smartest and most cash-rich entrepreneurs on Earth, using complex “off-balance-sheet financing” strategies to break down risks into many tradable segments, distributing them for different investors to digest.
• Even if it is not a bubble, it does not mean that all AI infrastructure investments can achieve a good ROI.

01 Understanding the Core: The Benefit Binding Mechanism of “Bundling”

The so-called “clustering” refers to the fact that this AI infrastructure tightly binds the interests of five parties:

• Tech giants (Meta, Microsoft, Google) and their large model partners (OpenAI, xAI): need computing power but don't want to spend a large sum at once.
• Chip supplier (Nvidia): Needs continuous large orders to support its valuation.
• Private equity funds (Blackstone, Blue Owl, Apollo): Need new asset classes to expand asset management scale and collect more management fees.
• Neocloud (CoreWeave, Nebius) and hybrid cloud service providers (Oracle Cloud Infrastructure): provide infrastructure and computing power, but also require large companies to leverage long-term contracts for financing.
• Institutional investors (pension funds, sovereign funds, traditional funds like BlackRock): require stable returns higher than government bonds.

And these five parties have formed a “community of shared interests”, for example:

• Nvidia prioritizes supply to CoreWeave while investing in its equity.
• Microsoft provides a long contract to CoreWeave while assisting with its financing
• Blackstone provides debt financing while raising funds from pension funds.
• Meta and Blue Owl jointly established an SPV to share risks.
• OpenAI and other large model vendors continue to raise the standards for model parameters, inference capabilities, and training scale, effectively increasing the overall industry's computational power demand threshold. Especially under the deep binding with Microsoft, this “technology outsourcing, pressure internalization” cooperation structure allows OpenAI to act as a catalyst for accelerating the global capital expenditure competition without spending money. It is not a funding party, but the actual curator driving the leverage increase across the board.

No one can be self-sufficient, and this is the essence of “banding together”.

02 Capital Structure — Who is Funding? Where is the Money Going?

To understand the overall architecture, we can start with the funding flow diagram below.

Tech giants need astronomical computing power, and there are two paths:

1. Self-built data center: This is the traditional model. The advantage is complete control, while the disadvantage is slow construction, and all capital expenditures and risks are borne on one's own balance sheet.
2. Seeking external supply: The giants are not simply renting servers, but have given rise to two core “external supplier” models. This is the new trend of the moment and is the focus of our analysis.

The first type is SPV (Special Purpose Vehicle) / Special Purpose Entity, which is a purely financial instrument. You can think of it as a special entity established specifically for “a single project, a single client.”

• Business Model: For example, Meta wants to build a data center but doesn't want to spend a large sum of money all at once, so it partners with an asset management company to form an SPV. The sole task of the SPV is to construct and operate this center exclusively for Meta. Investors receive high-quality debt instruments backed by rental cash flows (a mix of corporate bonds and project financing).
• Customer Type: Extremely Singular, usually only one (e.g., Meta).
• Risk level: Life and death completely depend on a single customer's credit.

The second type is Neocloud ( such as CoreWeave, Lambda, Nebius ), which are independent operating companies (Operating Company, OpCo) with their own operational strategies and full decision-making power.

• Business Model: For example, CoreWeave self-finances (equity and debt) to purchase a large number of GPUs, which are then leased to multiple clients under “minimum guarantee/reservation” contracts. Flexible but with significant fluctuations in equity value.
• Customer Type: Theoretically diverse, but in practice, it heavily relies on large companies in the early stages (e.g., Microsoft's early support for CoreWeave). Due to its smaller scale, unlike SPV which is backed by a single wealthy benefactor, Neocloud has a higher dependency on upstream suppliers (Nvidia).
• Risk level: Risks are spread across multiple clients, but operational capability, technology, and equity value all affect survival.

Although fundamentally different in legal and operational structures, the commercial essence of both converges: they are both “external suppliers of computing power” for the giants, removing massive GPU procurement and data center construction from the balance sheets of these giants.

So where does the money for these SPV and Neoclouds come from?

The answer is not traditional banks, but private credit funds. Why?

This is because after 2008, the Basel III Accord imposed strict requirements on banks' capital adequacy ratios. Banks taking on such high-risk, high-concentration, long-term large loans must set aside reserves that are excessively high and not cost-effective.

The business that banks “cannot do” or “dare not do” has created a huge vacuum. Private equity giants like Apollo, Blue Owl, and Blackstone have filled this gap—they are not restricted by banking regulations and can provide more flexible and faster financing, but at higher interest rates. They guarantee it with project rents or GPU/equipment with long-term contracts.

For them, this is an extremely attractive pie - many have traditional infrastructure financing experience, and this theme is enough to grow the scale of managed assets several times, greatly increasing management fees and carried interest (.

So where does the money for these private equity credit funds ultimately come from?

The answer is institutional investors (LPs), such as pension funds, sovereign wealth funds, insurance companies, and even general investors (for example, through the private credit ETF issued by BlackRock - which includes the 144A private debt Beignet Investor LLC 144A 6.581% 05/30/2049 under the Meta project).

The transmission path of the risk chain is thus established:

) ultimate risk bearer ( pension funds/ETF investors/sovereign funds → ) intermediary institutions ( private credit funds → ) financing entities ( SPV or Neocloud ) such as CoreWeave ( → (end users) tech giants ) such as Meta (

03 SPV Case Analysis — Meta's Hyperion

To understand the SPV model, Meta's “Hyperion” project is an excellent case study (with sufficient public information):

![])https://img-cdn.gateio.im/webp-social/moments-6fa1d9ac6cb54da881ce20b9b4c46e9b.webp(

• Structure/Equity: Meta and Blue Owl manage a fund group JV (Beignet Investor LLC). Meta holds 20% equity, Blue Owl holds 80%. Bonds are issued under SPV 144A structure. JV caps assets, and Meta leases them under a long-term agreement. Capital expenditures during the construction period are in the JV, and assets gradually transfer to Meta's balance sheet after financing leases begin.
• Scale: Approximately $27.3 billion in debt (144A private placement bonds) + approximately $2.5 billion in equity, making it one of the largest single corporate bond/private credit project financings in U.S. history. The maturity date is in 2049, and this long-term amortization structure essentially locks in the most challenging time risk first.
• Interest Rate/Rating: The debt has an S&P A+ rating (high rating allows insurers to allocate), with a coupon rate of approximately 6.58%.
• Investor structure: PIMCO subscribed 18 billion; BlackRock's ETFs totaled over 3 billion. For this group of investors, this is an extremely attractive high-quality stable return.
• Cash Flow and Lease: Blue Owl is not focused on the potentially depreciating GPUs (I think some people in the market are missing the point by worrying too much about the assumed long depreciation period of GPUs, because GPUs are just hardware, while the overall value of AI lies in the hardware + model. The price of older hardware declines due to iteration, but that doesn't mean the value of the final AI model application also decreases), but rather on the SPV cash flow supported by Meta's long-term lease (starting from 2029). The construction phase funds are also allocated to U.S. Treasury bonds to reduce risk. This structure combines corporate bond liquidity with project financing protection clauses, while being 144A-for-life (limited investor circle).

So why is the short-term risk of this architecture extremely low?

![])https://img-cdn.gateio.im/webp-social/moments-917994cc1a176f2068a5950cdbd77e2b.webp(

This is because under this structure, the Hyperion task is simple: the left hand collects Meta rent, while the right hand pays Blue Owl interest. As long as Meta does not collapse (the probability in the foreseeable future is extremely low), the cash flow remains as solid as a rock. There is no need to worry about fluctuations in AI demand or GPU price declines.

This 25-year ultra-long-term, rent amortization debt structure locks in all recent refinancing risks as long as rental income remains stable and interest is paid normally. This is the essence of “buying time” (allowing the value created by AI applications to gradually catch up with the financial structure).

At the same time, Meta uses its own credit and strong cash flow to secure massive long-term financing that bypasses traditional capital expenditures. Although under modern accounting standards (IFRS 16), long-term leases ultimately still appear on the balance sheet as “lease liabilities,” the advantage is that the pressure of capital expenditures amounting to billions of dollars during the initial construction phase, as well as the associated construction risks and financing operations, are first transferred to the SPV.

Transform a one-time massive capital expenditure into lease payments spread over the next 25 years, greatly optimizing cash flow. Then bet whether these AI investments can generate sufficient economic benefits in 10-20 years to cover principal and interest (considering a bond with a 6.58% coupon rate and operational costs, the ROI calculated based on EBITDA must be at least 9-10% to provide equity holders with a decent return rate).

04 Neocloud's Cushion — OpCo's Equity Risks

If the SPV model is “credit transfer”, then CoreWeave and Nebius, which are Neocloud models, represent “further layering of risk.”

![])https://img-cdn.gateio.im/webp-social/moments-26b2f96d0b55bc4a88329c635d75e565.webp(

Taking CoreWeave as an example, the capital structure is much more complex than that of an SPV. Multiple rounds of equity and debt financing involve investors such as Nvidia, VC, growth funds, and private debt funds, creating a clear risk buffer sequence.

What will happen if AI demand is not as expected, or new competitors emerge, and CoreWeave's income plummets and cannot pay high interest?

• The first step is the evaporation of equity value: CoreWeave's stock price plummets. This is the “equity buffer” - the first to absorb the shock. The company may be forced to raise funds at a discount, leading to significant dilution of existing shareholders' equity, or even total loss of investment. In contrast, the equity buffer for the SPV is thinner, as it cannot directly raise funds in the public market.
• The second step is that creditors suffer losses: only after the equity is completely “burned” will it be Blackstone and other private debt creditors' turn to bear the losses when CoreWeave is still unable to repay its debts. However, these funds usually require excellent collateral (latest GPUs) and strict repayment priority when lending.

CoreWeave and Nebius both adopt the strategy of “first securing long-term contracts, then financing against them,” which allows for rapid expansion through refinancing in the capital markets. The brilliance of this structure lies in the fact that large clients can achieve better capital utilization efficiency by leveraging future procurement contracts to drive more capital expenditure without having to invest upfront, thereby limiting the risk contagion to the entire financial system.

Conversely, Neocloud shareholders need to be aware that they occupy the most turbulent yet exhilarating position in this gamble. They are betting on high-speed growth while praying that the management's financial operations (debt extension, equity issuance) are nearly flawless, and they also need to pay attention to the debt maturity structure, pledge scope, contract renewal windows, and customer concentration in order to better assess the risk-reward ratio of equity.

We can also imagine that if the demand for AI grows slowly, who would be the marginal capacity most easily abandoned? SPV or Neocloud? Why?

05 Oracle Cloud: The Counterattack of an Atypical Cloud Player

While everyone is focusing on CoreWeave and the three major cloud giants, an unexpected “dark horse” in the cloud sector is quietly rising: Oracle Cloud.

It does not belong to Neocloud, nor is it part of the first-tier camp of the three major tech giants, yet it has secured contracts for part of the computing load of Cohere, xAI, and even OpenAI through highly flexible architecture design and deep cooperation with Nvidia.

Especially when the leverage of Neocloud becomes tighter and traditional cloud space is insufficient, Oracle, with its positioning of “neutral” and “replaceable,” becomes an important buffer layer in the second wave of AI computing power supply chain.

Its existence also shows us that this battle for computing power is not only a showdown among the three giants, but also that non-typical yet strategically significant suppliers like Oracle are quietly vying for position.

But don't forget, the table of this game is not just in Silicon Valley, but extends to the entire global financial market.

The government's “implicit guarantee” that everyone covets.

Finally, in this game dominated by tech giants and private finance, there is a potential “trump card” - the government. Although OpenAI recently stated that they “do not want” the government to provide loan guarantees for data centers, the discussion with the government is about potential guarantees for chip factories rather than data centers. However, I believe that their (or similar participants) original plan certainly included the option of “bringing the government in to form a coalition.”

How to say? If the scale of AI infrastructure becomes so large that even private placement debts cannot bear it, the only way out is to upgrade to a national power competition. Once the leading position of AI is defined as “national security” or “the moon landing competition of the 21st century”, government intervention becomes logical.

The most effective way to intervene is not to directly provide money, but to offer “guarantees.” This approach can bring a decisive benefit: significantly reducing financing costs.

Investors around my age should still remember Freddie Mac ) and Fannie Mae (. These two “Government Sponsored Enterprises” (GSEs) are not official departments of the U.S. government, but the market generally believes they have an “implicit government guarantee.”

They purchase mortgage loans from banks, package them into MBS and guarantee them, and after selling them on the open market, redirect the capital back to the mortgage loan market, increasing the funds available for lending. It is their existence that made the impact of the 2008 financial tsunami even greater.

Imagine if in the future, there is a “National AI Computing Company” backed by the government. The bonds it issues would be regarded as quasi-sovereign debt, with interest rates approaching those of U.S. Treasury bonds.

This will completely change the previously mentioned “buying time for productivity to rise”:

• Extremely low financing costs: The lower the borrowing costs, the lower the requirements for the “speed of AI productivity improvement.”
• Unlimited extension of time: More importantly, it allows for continual roll over at a very low cost, equivalent to buying nearly unlimited time.

In other words, this approach greatly reduces the chances of the gamble “blowing up” directly. However, once it blows up, the impact may expand by dozens of times.

06 Trillion Dollar Bet — The Real Key is “Productivity”

All the financial structures mentioned above - SPV, Neocloud, private debt - no matter how sophisticated, only answer the question of “how to pay.”

The fundamental question regarding whether AI infrastructure will become a bubble is: “Can AI truly increase productivity?” and “How fast?”

All financing arrangements lasting 10 or 15 years essentially “buy time.” Financial engineering gives the giants a breathing space, without the need for immediate results. But buying time comes at a cost: investors in Blue Owl and Blackstone (pension funds, sovereign funds, ETF holders) need stable interest returns, while equity investors in Neocloud need several times the valuation growth.

The “expected return rate” of these financing parties is the threshold that AI productivity must overcome. If the productivity gains brought by AI are unable to cover the high financing costs, this intricate structure will begin to collapse from its most vulnerable point (the “equity buffer”).

Therefore, in the coming years, special attention should be paid to the following two aspects:

• The speed of launching “application solutions” in various fields: Having a powerful model (LLM) is not enough. We need to see real “software” and “services” that can get enterprises to spend money. It is necessary for this type of application to be widely adopted, generating cash flows large enough to repay the principal and interest of the enormous infrastructure costs.
• Constraints of external limitations: AI data centers are energy-hungry monsters. Do we have enough power to support the exponential growth in computational demands? Is the upgrade speed of the power grid keeping pace? Will the supply of Nvidia's GPUs and other hardware encounter bottlenecks, making it “slower than” the timelines required by financial contracts? Supply-side risks could render all the “bought time” exhausted.

In short, this is a race between finance (cost of funding) and physical (electricity, hardware) and business (application landing).

We can also estimate roughly how much productivity improvement AI needs to bring to avoid a bubble in a quantitative way:

• According to Morgan Stanley's estimate, this round of AI investment should reach 3 trillion dollars by 2028.
• The aforementioned SPV bond issuance cost of Meta is about 6-7%, while according to a report by Fortune, CoreWeave's current average debt interest rate is around 9%. Assuming that the majority of private debt in the industry requires a return of 7-8% with a debt-to-equity ratio of 3:7, the ROI for these AI infrastructures (calculated with EBITDA and total capital expenditure) needs to be at 12-13% to allow for an equity return of over 20%.
• Therefore, required EBITDA = 3 trillion × 12% = 360 billion USD; if calculated based on an EBITDA profit margin of 65%, the corresponding revenue is approximately 550 billion USD;
• With an estimated GDP of about 29 trillion in the name of the United States, approximately 1.9% of GDP's additional output needs to be supported long-term by AI empowerment.

The threshold is not low, but it is not a fantasy either. The global cloud industry revenue in 2025 is expected to be around $400 billion. In other words, we need to see at least one to two cloud industries being revitalized by AI. The key lies in whether the speed of application monetization can be synchronized with the physical bottlenecks.

Risk Scenario Stress Testing: What happens when “time” is not enough?

All the financial structures mentioned above are betting that productivity can outpace financing costs. Let me use two stress tests to simulate the chain reaction when AI productivity realization speed is slower than expected:

In the first scenario, we assume that AI productivity is realized “slowly” (for example, achieving scalability in 15 years, but much of the financing may be for a 10-year term):

• Neocloud was the first to fall: Independent operators like CoreWeave, due to their high leverage, could not cover the high interest, resulting in their “equity buffer” being burned out, triggering debt defaults or discounted restructuring.
• SPVs face maturity risk: When SPV debts like Hyperion mature, Meta must decide whether to refinance at a higher interest rate (the market has already witnessed the failure of Neocloud), which erodes core business profits.
• Private credit fund LPs have suffered significant losses, and technology stock valuations have been substantially revised down. This is an “expensive failure,” but it will not trigger a systemic collapse.

In the second scenario, we assume that AI productivity has been “falsified” (technological progress has stagnated or costs cannot be reduced and scaled):

• Tech giants may opt for “strategic default”: this is the worst-case scenario. Giants like Meta may determine that “continuing to pay rent” is a bottomless pit, and subsequently choose to forcibly terminate leases, forcing SPV debt restructuring.
• SPV bond collapse: Bonds like Hyperion, which are considered A+ rated, will instantly decouple from Meta, causing prices to plummet.
• It could completely destroy the private credit “infrastructure financing” market and is very likely to trigger a crisis of confidence in the financial market through the aforementioned interconnectedness.

The purpose of these tests is to transform the vague question of “whether it is a bubble” into specific situational analyses.

07 Risk Thermometer: A Practical Observation Checklist for Investors

As for the changes in market confidence, I will continuously monitor five things as a risk thermometer:

• The speed of realization of AI project productivity: includes the acceleration or deceleration of expected revenue from large model vendors () linear growth or exponential growth), and the application status of different AI products and projects.
• Neocloud company's stock price, bond yield, announcements: including large orders, defaults/amendments, debt refinancing (some private placements will mature around 2030, requiring special attention), and capital increase pace.
• SPV bond's secondary price/yield spread: whether private placements like Hyperion maintain above par, whether trading is active, and whether ETF holdings are increasing.
• Changes in the quality of long-term agreements: take-or-pay ratio, minimum retention period, customer concentration, price adjustment mechanism (adjustments for electricity prices/interest rates/pricing related to inflation).
• Power progress and potential technological innovations: As the most likely external factor to become a bottleneck, attention needs to be paid to policy signals regarding substations, transmission and distribution, and electricity pricing mechanisms. Additionally, whether there are new technologies that can significantly reduce electricity consumption.

Why is this not a repeat of 2008?

Some people may draw comparisons to the bubble of 2008. I believe this approach may lead to misjudgments:

The first point lies in the different nature of core assets: AI vs. housing.

The core asset of the 2008 subprime mortgage crisis was “housing.” The houses themselves do not contribute to productivity (rental income growth is very slow). When housing prices deviate from the fundamentals of residents' income and are packaged into complex financial derivatives, it is only a matter of time before the bubble bursts.

The core asset of AI is “computing power.” Computing power is the “production tool” of the digital age. As long as you believe that AI is highly likely to substantially increase the productivity of the entire society (software development, drug research and development, customer service, content creation) at some point in the future, you don't have to worry too much. This is a “prepayment” for future productivity. It has real fundamentals as a anchor point, but it has not yet been fully realized.

The second point is that the key nodes of the financial structure are different: direct financing vs. banks.

The 2008 bubble spread significantly through key nodes (banks). Risks were transmitted via “indirect financing through banks.” The collapse of one bank (like Lehman) triggered a crisis of trust in all banks, leading to a freeze in the interbank market and ultimately igniting a systemic financial crisis that affected everyone (including a liquidity crisis).

Currently, the financing structure of AI infrastructure is mainly based on “direct financing”. If AI productivity is discredited, CoreWeave goes bankrupt, and Blackstone defaults on its $7.5 billion debt, this will lead to significant losses for Blackstone investors (pension funds).

The banking system has indeed become stronger since 2008, but we cannot oversimplify and think that risks can be completely “contained” in the private equity market. For example, private credit funds themselves may also leverage bank funding to amplify returns. If AI investments generally fail, these funds could still suffer significant losses that may spill over through two pathways:

1. Leverage default: The fund's leverage financing default to the bank will transmit the risk back to the banking system.
2. LPs Impact: Pension funds and insurance companies face deteriorating balance sheets due to significant investment losses, prompting them to sell off other assets in the public market, triggering a chain reaction.

Therefore, a more accurate statement is: “This is not the kind of interbank liquidity crisis triggered by a single point explosion and a complete freeze in 2008.” The worst-case scenario would be “expensive failures,” with lower contagion and slower speed. However, due to the opacity of the private equity market, we must remain highly vigilant about this new type of slow contagion risk.

Insights for investors: At which layer of this system are you?

Let's return to the original question: Is AI infrastructure a bubble?

The formation and bursting of bubbles come from the huge gap between expected benefits and actual results. I believe that in general, it is not a bubble, but rather a precise high-leverage financial layout. However, from a risk perspective, in addition to certain aspects that need special attention, we cannot be complacent about the “negative wealth effect” that small-scale bubbles may bring.

For investors, in this trillion-dollar AI infrastructure race, you must know what you are betting on when holding different assets:

• Tech giant stocks: You're betting that AI productivity can outperform financing costs
• Private Credit: You earn stable interest, but bear the risk of “time may not be enough.”
• Neocloud equity: You are the first buffer with the highest risk and highest reward.

In this game, position determines everything. Understanding this series of financial structures is the first step to finding your own position. And understanding who is “curating” this show is key to determining when this game will end.

Source: Distill AI