OpenAI's $122B round: $37B actual cash, rest is vendor deals and contingent capital

The headline number is $122 billion. What hit the bank account on day one is closer to $37 billion. Amazon's $50B includes $35B contingent on IPO or AGI. Nvidia's $30B is compute credits, not cash. SoftBank's $30B arrives in tranches. The actual VC-style capital is $12 billion. If you are selling into enterprise AI, this structure matters: it signals how vendor partnerships are replacing traditional funding in late-stage AI deals.

OpenAI's $122B round: $37B actual cash, rest is vendor deals and contingent capital

The Structure Is the Story

OpenAI closed $122 billion at an $852 billion post-money valuation on 31 March 2026. The headline is massive. The actual cash position is different.

Here is what actually happened.

What Hit the Bank Account: $37B

Of the $122 billion announced:

  • Amazon committed $50B, but only $15B is immediate. The other $35B is contingent on OpenAI going public or achieving AGI by year end.
  • SoftBank committed $30B in three quarterly tranches. First tranche: $10B at close.
  • Nvidia committed $30B, but it is compute capacity, not cash. Dedicated inference and training systems. OpenAI's capex flows back to Nvidia as GPU purchases.
  • The remaining $12B came from a16z, D.E. Shaw, TPG, T. Rowe Price, Microsoft, and $3B from retail investors via banks and ARK ETFs.

Actual immediate capital: roughly $37 billion. The rest is conditional, deferred, or vendor credits.

The Vendor Deal Layer

Amazon's investment includes a separate agreement for OpenAI to spend $100 billion on AWS infrastructure over eight years. Amazon invests $50B, OpenAI commits to spend $100B back on AWS. That is a customer contract structured as a funding round.

Nvidia's $30B is chips and infrastructure Nvidia is supplying. OpenAI trains on Nvidia systems. The capital flows back to Nvidia as revenue.

SoftBank has committed $3 billion annually to deploy OpenAI technology across its portfolio companies. Investor and customer, simultaneously.

Why This Matters for Sales Teams

If you are selling into enterprise AI or partnering with AI vendors:

  1. Vendor partnerships are replacing traditional funding. Late-stage AI deals are increasingly structured as strategic vendor relationships, not pure equity bets.
  2. Revenue commitments matter more than valuation headlines. OpenAI's $100B AWS spend is more concrete than the $852B valuation.
  3. Enterprise adoption signals are in the deal structure. SoftBank's $3B annual deployment commitment across portfolio companies is a go-to-market play, not just an investment.

The Guaranteed Return Footnote

Separate from the $122B round, OpenAI is negotiating a $10B pre-money joint venture with TPG, Advent, Bain Capital, and Brookfield. The terms: a guaranteed minimum return of 17.5% plus early model access.

That is unusual. Preferred equity comes with priority returns, but a guaranteed floor of 17.5% is not standard, even in PE.

The play: PE firms collectively own hundreds of operating companies. Early model access means guaranteed enterprise adoption at scale.

If you are calling on portfolio companies of those PE firms, expect OpenAI to already be in the building.

The Honest Breakdown

  • $37B: actual cash at close
  • $35B: contingent on IPO or AGI milestone
  • $30B: Nvidia compute credits
  • $20B: SoftBank tranches due later in 2026

Total announced: $122B. Total immediate capital: $37B.

The structure is the story. For sales teams, the vendor partnership layer is the real signal: AI companies are securing distribution and infrastructure through funding rounds, not just raising capital.