SemiAnalysis: Not Bearish on Nvidia; The 'AI Central Bank' Could Move a $7 Trillion Debt Snowball

Original Author: Zhao Ying

Original Source: Wall Street News

On July 6th, renowned semiconductor research firm SemiAnalysis posted six consecutive tweets on X platform, revealing that Nvidia's Kyber NVL144 rack is delayed by over 12 months due to PCB mid-plane manufacturing challenges. Asian AI hardware supply chain stocks plummeted in response.

Nvidia subsequently stated that the roadmap remains unchanged, but did not disclose specific progress details.

The controversy is far from settled. On July 7th, SemiAnalysis published another lengthy paid article, aiming its criticism at Nvidia. However, this time, the firm did not play the role of a "bear."

SemiAnalysis projects that by 2029, the global scale of AI debt financing will exceed $7 trillion. What does $7 trillion represent? It would become the second-largest asset-backed debt market globally, second only to the U.S. residential mortgage market (approximately $13 trillion).

What is Nvidia's role in this? SemiAnalysis disclosed a strategic move by Nvidia—the "backstop" plan. Nvidia is leveraging its own AA/Aa2 investment-grade credit rating to provide minimum revenue guarantees for AI compute rental companies, thereby encouraging banks to lend. In other words, Nvidia is acting as the lender of last resort and insurer for the entire AI ecosystem, booking large sales while transferring some of the risk of downstream demand shortfalls onto itself. SemiAnalysis likens Nvidia to the "central bank of AI."

Regarding the discussions on X platform about whether SemiAnalysis is "bearish on Nvidia," the firm stated:

The firm has not published any positive or negative views on Nvidia stock; it is merely accurately capturing supply chain and technical details. The market can trade based on its own analysis.

The AI Debt Snowball: Reaching $7 Trillion by 2029, Nearing the U.S. Mortgage Market

SemiAnalysis believes that AI infrastructure construction is forming a multi-trillion dollar credit market. By 2029, outstanding AI-related debt could reach approximately $7.1 trillion, surpassing the scale of all other U.S. asset-backed debt markets except for the mortgage financing market.

This debt primarily comes from two types of capital expenditure. One is AI IT capital expenditure, including GPUs, networking, storage, and supporting CPUs; the other is AI data center capital expenditure, including the server rooms, electricity, and cooling infrastructure required to house these GPUs.

Previously, cloud giants like Google, Amazon, Meta, Microsoft, and Oracle primarily relied on their internal cash flow to build AI clusters. However, in the past year, Oracle, Meta, and even Google have started to use more debt. As project scales continue to expand, the market constraint is no longer just about securing GPUs or finding server rooms, but about whether they can borrow enough cheap, long-term capital.

SemiAnalysis concludes that the financing methods for AI capital expenditure are changing. The balance sheets of cloud giants are not infinite. If all AI clusters rely on endorsements from a few investment-grade cloud vendors, new projects will inevitably encounter credit bottlenecks sooner or later.

The "Trilemma": Capital, Customers, and Data Centers are all Indispensable

SemiAnalysis breaks down AI project financing into a "trinity": capital, underwriting contracts, and data centers.

First is capital. Lenders typically require long-term take-or-pay contracts from investment-grade cloud vendors, or similar credit guarantees, before they are willing to lend. In other words, lenders truly value the creditworthiness of the backend customer, not the Neocloud itself.

Second is underwriting. To secure customers, Neoclouds often need to first prove they can pay GPU deposits and lock in equipment. However, to secure equity funding, they need to first prove they have customers and loans. This creates a chicken-and-egg cycle in the early stages of a project.

Third is the data center. Neoclouds must either use customer contracts and financing to convince data center operators to lease capacity, or build their own data centers. The latter option carries greater financial pressure and longer lead times.

This model locks the market into a "5-year, cloud giant-endorsed" template. The problem is that many VC-backed AI startups and inference service providers need short-term, large-scale compute power, not 5-year long-term contracts. Inference providers are particularly reluctant to bear long-term price and demand risks, and in many cases, they would rather forgo compute power than sign a lease longer than one year.

Nvidia, the "AI Central Bank": Leveraging AA-rated Credit to Move the Market

Nvidia proposed the "Backstop Plan" to bridge this financing gap.

According to SemiAnalysis, Nvidia provides a revenue floor for GPU rentals to Neoclouds. If third-party customer demand is insufficient, Nvidia commits to purchasing the compute capacity at a pre-set price; if the Neocloud rents out the compute at a higher price, Nvidia shares in a portion of the excess revenue.

These arrangements are typically for 6-year terms, providing a minimum revenue guarantee on the underlying GPU capacity based on a pre-agreed price curve. The Neocloud can still lease the compute power to any customer and offer more flexible lease terms. Nvidia's guarantee is only triggered when market demand is insufficient and the capacity cannot be leased at market prices.

This is the origin of the "AI central bank" analogy. Nvidia isn't literally issuing currency, but it is playing a role akin to the buyer of last resort and credit underwriter within the AI compute credit system. Lenders can use Nvidia's AA/Aa2 credit rating to assess the worst-case scenario for a project, making them more willing to lend.

For Nvidia, this helps expand the buyer base for its GPUs. If the market were limited to a few large cloud vendors signing 5-year underwriting contracts, GPU demand would quickly hit financing constraints. Moreover, these cloud vendors are developing their own chips to compete with Nvidia's systems. Supporting Neoclouds and more enterprise clients effectively opens new financing channels for GPU demand.

Deconstructing the Backstop Plan: How Much Does Nvidia Earn vs. the Neocloud?

SemiAnalysis emphasizes that Neoclouds don't use Nvidia's credit for free. Under the backstop structure, Neoclouds sacrifice some upside potential in exchange for project bankability.

In an example price curve, the 6-year average backstop price is $2.36 per GPU per hour. Assuming a GB300 one-year lease price of $6.75 per hour in the first year, and a first-year backstop price of $3.68 per hour, the difference between the customer price and the backstop price is $3.07. If Nvidia takes 40% of the amount exceeding the backstop price, Nvidia gets $1.23, the Neocloud gets $1.84, and the Neocloud's actual first-year revenue is $5.52 per hour, lower than the $6.75 without the backstop.

Over six years in this scenario, Nvidia's average fee is about 18%. The Neocloud's project IRR also decreases. In the scenario with Nvidia's backstop and primarily 1-year short-term leases, the project IRR is 25.4%; without the backstop but still able to secure financing and lease out capacity, the IRR could reach 40.7%.

The key is the worst-case scenario. If demand is weak and the Neocloud can only lease the compute back to Nvidia, the project returns could be near zero or even slightly negative. Lenders don't require the project to be profitable in the worst case, only that it can still service its debt. Therefore, the viability of the debt increasingly depends on the reliability of Nvidia's backstop.

This is the core point investors should focus on: Nvidia's arrangement helps drive GPU sales and Neocloud expansion in the short term. However, if compute demand falls short of expectations, the revenue shortfall will be borne by Nvidia. The debt may not be directly on Nvidia's balance sheet, but the safety net for the financing model is increasingly concentrated on Nvidia's credit.

GPU Financing Pricing Ultimately Depends on Who is Backing It

SemiAnalysis states that current pricing in the GPU financing market is determined less by the Neocloud's own credit and more by who signs the long-term underwriting contract.

CoreWeave serves as a benchmark. Its 5-year unsecured bond yield is around 10%. However, in its $8.5 billion DDTL 4.0 delayed draw term loan backed by Meta, the fixed-rate portion costs approximately 5.9%, only 90 basis points higher than Meta's 5-year bond yield of about 5.0%. This 90 basis point spread roughly reflects the market's pricing for CoreWeave's execution risk.

If a Neocloud operates without a long-term cloud vendor underwriting, its financing costs increase significantly. For top-tier Neoclouds, unsecured financing might require paying an interest rate of about 10%, roughly 4 percentage points higher than financing with a backstop. At a 70% to 80% loan-to-value ratio, financing costs rise from 5.62% to 10%, causing the pre-tax profit margin to drop from 14.8% to 5.4%.

Nvidia's backstop positions the pricing between these two benchmarks: higher than the total yield of around 5.9% for current cloud vendor-backed deals, but lower than the ~10% yield on CoreWeave's unsecured bonds. Banks focus most on the debt service coverage ratio (DSCR). For projects with Nvidia's backstop, loan sizes are typically calculated based on a scenario where the backstop is triggered, requiring a DSCR of at least 1.3x in the early years, corresponding to a loan-to-value ratio usually between 70% and 80%.

Public Projects Expand in Asia-Pacific, Backstop Model Begins to Materialize

Publicly disclosed Nvidia backstop projects are currently concentrated in the Asia-Pacific region.

The first is SharonAI's 72MW AI factory in Australia. Announced in June 2026, the project plans to expand to up to 40,000 GB300 GPUs under a 6-year backstop. SharonAI disclosed a total backstop value of $4.88 billion, translating to an average floor price of about $2.33 per GPU per hour over six years.

Another is Firmus's 360MW AI cluster in Batam, Indonesia, potentially located within DayOne's facility at the Kabil Industrial Tech Park. Announced on June 29, 2026, this indicates Nvidia's backstop is entering a larger scale.

Firmus expects six-year customer revenue from this project to be between $25 billion and $30 billion, targeting AI-native companies, enterprise clients, and inference providers, offering various lease terms. However, before deploying the GPUs, Firmus still needs to secure a data center provider or continue with self-building.

SemiAnalysis also notes that Nvidia is not the only GPU maker using backstop arrangements. AMD has offered similar arrangements to clients like AWS, OCI, DigitalOcean, Vultr, Tensorwave, and Crusoe since last year: clients purchase more AMD GPUs, and AMD agrees to potentially lease back a portion under a long-term contract for internal software development if the Neocloud cannot fully sell the capacity.

SemiAnalysis Denies Bearish Stance, but Market is More Sensitive to its Signals

The publication of this article comes amidst controversy surrounding SemiAnalysis itself.

On the morning of July 6th, SemiAnalysis posted a series of tweets on X, claiming Nvidia's Kyber NVL144 rack architecture faced significant delays, pushing it back over 12 months to 2028. The news caught attention before market open and caused declines in multiple AI hardware supply chain stocks in Japan, South Korea, and the Taiwan region. Nvidia subsequently responded that its product roadmap remains unchanged, denying that the core progress was affected.

This made SemiAnalysis's subsequent article more susceptible to being interpreted by the market as either bearish or bullish on Nvidia. In response, SemiAnalysis stated on X that they had not published any positive or negative views on Nvidia stock, merely sharing the company's supply chain and technical details.

Crackerjack Finance pushed back against the "bearish" interpretation, noting that SemiAnalysis's charts show actual data for the second half of the year is 20% higher than market expectations, and extrapolated an earnings per share of around $15 for the next fiscal year, suggesting a stock price in the $300-$400 range. THE Grand Poobah commented that "tri-party circular financing seems insufficient," pointing to market concerns over the increasing complexity of financing structures.

The core issue is that AI-related assets have experienced years of appreciation, with valuations and expectations sitting at high levels. Any signals of supply chain risk or changes in financing structures can be rapidly amplified. While SemiAnalysis's clarification indicates it did not directly offer a stock opinion, following the Kyber NVL144 incident, the market influence and credibility debates surrounding its supply chain revelations will continue to coexist.

For investors, the true meaning of this "lengthy report" is that the AI race is no longer just about "who has the GPUs," but about "who can simultaneously assemble GPUs, debt, customer contracts, and data centers." Nvidia's backstop mechanism could continue to amplify GPU demand, but it also risks concentrating more of the tail-end pressure of the AI debt cycle onto Nvidia's own creditworthiness.