SpaceX Wants $40 Billion to Buy Nvidia Chips: Why the AI Boom Is Becoming a Financing Boom
By Generational Wealth Investments | GenerationalWealth.biz
SpaceX completed the largest IPO in history in June. Less than 4 months later, it is reportedly back in front of lenders asking for another $40 billion, and the shopping list has a single line on it: Nvidia AI chips.
At Generational Wealth Investments, we don't chase hype, we decode the market. And the part of this story worth decoding is the financing, more than the chips.
For most of the past 3 years, the question hanging over the AI trade was supply: could anyone get enough GPUs? That question is fading. The new one is whether buyers can borrow enough to pay for them, and what lenders will charge. When a build-out stops being funded from cash flow and starts being funded from credit markets, the number that matters most moves from the earnings release to the bond term sheet. That is the thesis of this post: the AI boom is becoming a financing boom, and the cost of debt is now a leading indicator for the entire trade.
What the Financial Times Reported
According to a Financial Times report published Tuesday, October 6, and relayed by Reuters, SpaceX is seeking roughly $40 billion to fund a purchase of Nvidia AI chips. The financing has 2 pieces:
About $10 billion in bank loans
About $30 billion in investment-grade debt
Apollo Global Management is expected to lead the financing and help place the debt with a wide range of investors. Bond giant Pimco is reportedly among a small group of lenders in talks. The transaction is expected to close in 2027.
2 caveats belong up front. First, this is a report citing people familiar with the matter, not a company announcement, and Reuters said SpaceX, Apollo and Nvidia did not immediately respond to requests for comment. Second, a deal that closes in 2027 is still a negotiation. Size, pricing and structure can all change between now and then.
The market's first reaction was small but telling. SpaceX shares (SPCX) slipped about 1% in extended trading after the report, while Nvidia (NVDA) rose about 0.5%.
Why Borrow When You Just Raised a Record IPO?
SpaceX reported close to $101 billion in cash shortly after its June listing and carries a market value of around $2.17 trillion. Against that, $40 billion is less than 2% of the company's equity value. So why not just write the check? There are 3 reasons.
The spending plan is bigger than the cash pile. Elon Musk has said SpaceX will build its data centers exclusively with Nvidia hardware, and that the Colossus 2 facility could more than double its Nvidia chip count by December. The company has talked about ending 2026 with more than 2 gigawatts of computing capacity and approaching 10 gigawatts in 2027. Analyst estimates cited in recent coverage put capital spending in the hundreds of billions of dollars over the next several years. A $101 billion cushion does not cover that.
Debt doesn't dilute. Selling more stock hands a slice of the company to new owners and shrinks every existing holder's share, including Musk's. Borrowing keeps ownership intact. For a company with a premium valuation and an investment-grade rating, bonds are usually the cheaper form of capital, as long as lenders agree.
The business is not yet paying for itself. SpaceX reported a net loss of roughly $5.5 billion in the first half of 2026, with the AI unit as the main drag. A company that is losing money while spending at this pace can either slow down or find outside capital. SpaceX is choosing outside capital.
This is also not SpaceX's first trip to the bond market. Days after the IPO, it sold its first investment-grade bonds in a deal reported at $20 billion or more, with proceeds aimed at refinancing a bridge loan. A $40 billion follow-up earmarked for a single vendor is a different kind of borrowing. The June deal tidied up the balance sheet. This one would expand it.
The Circle: Nvidia Sits on Every Side of the Table
Nvidia is the supplier in this transaction. It is also a SpaceX shareholder, having disclosed a stake of nearly 123 million shares. And in August, Nvidia announced partnerships with Apollo, BlackRock, Blackstone, Brookfield, Goldman Sachs and KKR on financing platforms designed to mobilize more than $500 billion for AI infrastructure. That capital exists, in large part, to help customers buy Nvidia systems.
Apollo, the firm expected to lead SpaceX's financing, is one of those partners. It has also financed Nvidia GPU purchases for Musk's AI business before.
Nothing in the reporting says this specific deal runs through Nvidia's financing platforms. But the pattern is hard to miss: the same small group of institutions is supplying the chips, owning the equity and arranging the debt. In good times, that loop is efficient, because capital moves quickly to where demand is. In tougher times, stress in 1 link travels to the others faster than investors expect.
Why the 2 Stocks Moved in Opposite Directions
A 1% dip and a 0.5% gain are not big moves. The direction is the lesson.
For Nvidia, a financed order is a firmer order. Revenue that depends on a customer's future cash flow is uncertain. Revenue backed by $40 billion of committed lending is much closer to booked. Nvidia shareholders are being told, in effect, that a large piece of 2027 demand has a funding source attached.
For SpaceX, the same news is an obligation. Debt adds fixed interest payments and a repayment date to a business that is still reporting losses. Equity holders stand behind lenders in line, and every dollar of interest has to be covered before shareholders see anything.
So the same headline was read 2 ways: as demand visibility for the seller and as leverage for the buyer. That split tends to widen as borrowing grows.
The Mechanism: How a Credit Spread Becomes a Chip Order
Investment-grade bonds are priced as a spread over Treasury yields. That spread is the extra return lenders demand for taking on a particular borrower's risk. When demand for a deal is strong, the spread tightens. When lenders are nervous, it widens.
Now run the math. On $40 billion, every 1 percentage point of additional yield costs $400 million a year in interest. That flows straight into the cost of each data center, and from there into the return on every hour of computing SpaceX sells.
The chain works like this:
Lenders demand a wider spread.
The cost of each new gigawatt of capacity rises.
Projects at the margin stop clearing their return hurdle.
Buyers stretch out or trim their chip orders.
Chipmaker revenue forecasts come down.
There is a second issue inside the structure, and it is timing. AI chips have a useful life measured in a handful of years before newer generations make them less competitive. Bonds can run far longer. A borrower financing short-lived hardware with long-dated debt is betting that the revenue arrives before the hardware ages out.
SpaceX has real contracts supporting that bet, including multi-year computing agreements with Google and Anthropic worth tens of billions of dollars. The FT also put SpaceX's credit rating at BBB, which is investment grade but in its lower tier. Lenders will be weighing those contracts against the pace of spending and the losses.
Morgan Stanley estimates AI infrastructure will need about $1.5 trillion of outside financing by 2028. With that much debt looking for a home, the price lenders charge sets the pace of the whole build-out.
3 Scenarios for How This Plays Out
Scenario 1: Clean execution. The deal is heavily oversubscribed and prices close to where SpaceX's existing bonds trade. That tells you credit markets still have plenty of appetite for AI risk. More borrowers follow, and the market treats debt-funded chip orders as normal. Nvidia's own $25 billion bond sale in June, which drew orders of more than 3 times its size, is the benchmark for what strong demand looks like.
Scenario 2: Done, but at a price. The deal gets placed, but lenders extract concessions: a noticeably wider spread, shorter maturities, or collateral tied to the chips themselves. The build-out continues, but the cost of capital steps up for everyone behind SpaceX in line, and weaker balance sheets feel it first. The gap between how the market treats chip sellers and chip buyers keeps widening.
Scenario 3: Resized or restructured. The deal is cut, delayed past 2027, or moved into a more complex private structure because public demand isn't there at an acceptable price. That would be the first hard evidence that credit supply, rather than chip supply, is the limit on AI spending. It would force a rethink of 2027 and 2028 hardware forecasts across the sector.
We are not assigning odds. A scenario framework lets you decide in advance what each outcome means, so you aren't interpreting the news in a hurry when it arrives.
What Would Prove This Thesis Wrong
A thesis that can't be falsified is just a narrative. Here is what would tell us financing is not the governing variable:
Spreads stay flat as issuance balloons. If hundreds of billions of dollars in AI debt keep clearing without lenders asking for more yield, capital is abundant and the constraint is somewhere else.
Buyers fund from cash flow instead. If SpaceX's contracted revenue ramps quickly enough that the company shrinks or cancels the raise, the financing angle fades.
Supply bottlenecks return. If chip lead times lengthen and committed capital sits idle waiting for hardware or power, the limit is physical, not financial.
The Watchlist
Pricing: the spread on the $30 billion bond portion versus Treasuries and versus SpaceX's June bonds
Demand: how many times the order book covers the deal
Structure: whether the debt is secured by the chips, and how maturities compare with hardware life
Participants: whether Pimco and other large lenders commit, and on what terms
Ratings: any change in tone from Moody's, S&P or Fitch as SpaceX's borrowing climbs
Price action: whether SPCX and NVDA keep reacting in opposite directions to financing headlines
What This Means for You
You don't need to trade bonds to use this. The takeaway is about where to look. For 3 years, the AI story was told through revenue growth and chip shipments. From here, a meaningful part of it will be told through loan terms, order books and credit ratings. Those get far less attention on social media, and they tend to move first.
How you spend your million-dollar hours matters. An hour spent understanding who is financing a boom, and at what cost, is worth more than an hour refreshing a price chart.
Join the Generational Wealth Community
This is the Generational Wealth Community, your pathway from knowledge to legacy. We decode what moves markets. Follow or subscribe for the market setup every morning, and join the conversation at GenerationalWealth.biz/community.
⚠️ Educational Disclaimer: This content is produced by Generational Wealth Investments for educational and informational purposes only. Nothing here constitutes financial or investment advice. Markets are volatile. Always do your own research and consult a licensed financial professional before making investment decisions.
