Key Takeaways

  • SpaceX is pursuing $40 billion in debt financing, split into $10 billion in bank loans and $30 billion in investment-grade debt led by Apollo and PIMCO.
  • The entire capital pool is designated to buy Nvidia Vera Rubin chips, bypassing Tesla's custom in-house silicon efforts for raw speed.
  • Prospective lenders received a two-page deal memo showing pictures of outer space with an arrow claiming SpaceX will build data centers across the universe.
  • Private credit giants can underwrite this transaction because hardware acts as hard collateral and Elon Musk possesses an unmatched execution track record.

The Two-Page Pitch for $40 Billion

Most founders spend months building ninety-slide decks with five-year financial models to raise a seed round. SpaceX approached Wall Street for $40 billion with two pieces of paper.

According to Jordi Hays, prospective lenders received a document stripped of conventional underwriting tables. Instead, the memo contained “pictures of outer space and an arrow pointing out that the company was going to build data centers somewhere in the universe.”

It sounds like a parody of late-stage tech exuberance. Yet Apollo and PIMCO stepped up to anchor $30 billion in investment-grade debt, alongside $10 billion in bank facilities. The reason is simple: Musk has spent twenty years turning impossible physical claims into operational cash flows. As Hays observed, “giving money to Elon and just sort of trusting the process has worked to date. And so he's sort of earned the right to have a you know shorter than than average memo I think at this point.”

When you establish absolute credibility on execution, the friction of capital formation drops to zero. Brevity signals complete confidence.

Why Custom Silicon Takes a Backseat to Jensen

The capital destination is just as revealing as the deal size. Tesla designs proprietary silicon, yet SpaceX is routing this entire haul straight to Nvidia.

John Coogan pointed to the explicit commitment: “We've decided to build exclusively on Nvidia because we think the Vera Rubin architecture is the best architecture. We think it's the best AI computer and we greatly value our close cooperation and partnership on many levels.”

Founders often fall in love with vertical integration before they need it. Building custom hardware takes years of architectural iteration, compiler design, and foundry allocation. In a race where frontier compute dictates capability, waiting for your own chip team to catch up is suicide. Musk knows how to run custom silicon programs, but he is choosing to write a massive check to Jensen Huang instead. Speed beats theoretical cost efficiency when the prize is total category domination.

How Wall Street Underwrites Bare Silicon

Financing compute through debt changes how technology businesses scale. Software historically demanded equity because developers and customer acquisition cannot be seized in a liquidation. Hardware flips that equation.

As Coogan noted, “I was I was laughing at this too and then I was thinking like it really is pretty simple. Like it's $40 billion. Buy some chips. What is the value of the chips? How do you underwrite those?”

Credit firms like Apollo and PIMCO treat advanced GPUs as physical assets with secondary market value. If a company defaults, the lender takes the servers, sells compute contracts, or offloads the processors to rival data centers. For builders, this marks a shift in how infrastructure gets funded. When capital expenditures produce tangible, revenue-generating machines, debt preserves your equity.

What to Do With This

Audit your current fundraise or investor update and delete every slide that explains market size through generic industry graphs. Replace them with two pages: one showing the hard asset or technical moat you are buying, and one showing your past delivery record against stated milestones.