OpenAI’s $110 Billion Round Grew to $122 Billion—Then Musk Lost at Trial
OpenAI’s financing ultimately became larger than the deal first presented in February. In its March 31 closing notice, the company put the final round at $122 billion in committed capital and its post-money valuation at $852 billion—$12 billion above the original $110 billion raise.
The other major change came in court. On May 18, a federal judge dismissed Elon Musk’s claims after a nine-person advisory jury found that he had waited too long to sue; Musk then said he would appeal, according to the Associated Press account of the decision. That statement established his intention, but it did not itself establish a later appellate ruling.
Why the financing changed from $110 billion to $122 billion
The two funding figures describe different stages of the same transaction. OpenAI’s February 27 financing statement set out $110 billion in new investment at a $730 billion pre-money valuation: $50 billion from Amazon and $30 billion each from Nvidia and SoftBank. It also left room for additional financial investors to join while the round remained open.
A pre-money valuation measures the negotiated equity value before the new capital is added. On the original terms, adding the planned financing produced an implied post-money value of $840 billion. The subsequent $12 billion increase in committed capital brought that calculation to the final $852 billion figure.
The wording “committed capital” is important. It supports the conclusion that investors agreed to provide an exceptionally large pool of financing, but it does not mean the entire amount became unrestricted cash available for immediate spending on the closing date. Investment commitments can arrive in stages or remain subject to contractual closing mechanics.
The round also should not be read like a public-company market capitalization. Its valuation was negotiated in a private financing rather than established through continuous exchange trading, and the securities involved may carry rights or protections that ordinary public shares would not have. The price therefore shows what participating investors accepted under those terms, not what a liquid public market would necessarily pay.
The investors are tied to OpenAI’s infrastructure spending
This was not simply a transfer of capital from passive financial backers. Amazon and Nvidia also sell the cloud capacity and chips that OpenAI needs to train models and operate products, giving them commercial relationships with the company alongside their investment exposure. OpenAI, in turn, gains access to infrastructure while taking on substantial long-term purchasing commitments.
Amazon’s published partnership terms divide its $50 billion investment into an initial $15 billion and a further $35 billion after specified conditions are met. The agreement also makes AWS the exclusive third-party cloud distributor for OpenAI Frontier and expands the companies’ cloud arrangement by $100 billion over eight years, including approximately two gigawatts of Trainium capacity.
The conditions attached to the second Amazon tranche add uncertainty to the headline total. Bloomberg’s account of the completed financing tied that $35 billion to either an OpenAI initial public offering or the company reaching artificial general intelligence. An IPO is a recognizable legal and financial event; AGI depends on a contractual definition of a technological threshold that has no universally accepted test.
The overlap between investor and supplier can align incentives. Infrastructure providers benefit if OpenAI’s demand grows, while OpenAI obtains capacity that may otherwise be difficult to secure quickly. It also creates concentration and execution risks because part of the investment case depends on the company turning expensive computing commitments into products that generate durable revenue.
The round is evidence of investor demand, not proof against an AI bubble
The financing demonstrates that large strategic investors were willing to back OpenAI at an extraordinary private valuation. It does not establish that future revenue, margins or cash generation will justify that price. Those outcomes depend on customer demand, competitive pricing, infrastructure costs and the company’s ability to convert technical capability into products people and businesses continue paying for.
The optimistic interpretation is that more compute can support stronger models, wider distribution and greater enterprise adoption. The skeptical interpretation is that rapidly expanding infrastructure could lock the company into heavy costs while model competition pushes prices down. Both conclusions remain hypotheses; the completed round confirms access to capital, not the eventual return on that capital.
For the same reason, the transaction cannot by itself prove that artificial-intelligence valuations form a bubble. Bubble claims require evidence about prices relative to realistic future cash flows across a market, not simply the size of one private financing. Conversely, a successful fundraise is not proof that OpenAI’s economics are already secure.
The most consequential financial questions now concern deployment rather than fundraising publicity: how quickly committed capital becomes usable, whether conditional tranches arrive, and whether revenue can grow without infrastructure spending rising just as fast. An eventual public offering would expose the private valuation to a broader and more liquid market, but the funding agreement does not guarantee either an IPO or its price.
Musk’s loss ended the immediate trial threat, not the underlying dispute
Musk’s case accused OpenAI and its leaders of departing from the nonprofit purpose around which the organization was founded. The court did not resolve that broader historical dispute on its merits. The claims failed because the jury and judge concluded that Musk had missed the applicable filing deadline.
That distinction matters financially and legally. OpenAI avoided an adverse trial judgment and the remedies Musk had sought, removing an immediate source of uncertainty shortly after the larger financing closed. The result was not a judicial endorsement of every decision made during OpenAI’s organizational and commercial evolution.
The verified sequence is therefore narrower than the original speculation suggested. The preliminary funding commitments expanded before the round closed, and OpenAI then prevailed at the trial level because Musk’s claims were untimely. The remaining investment question is whether the company can earn returns commensurate with its valuation and infrastructure obligations—not whether the size of the round alone settles the debate over AI’s economics.
Also read:
Subscribe to our newsletter
Get the latest Web3, AI, and crypto news delivered straight to your inbox.