Why OpenAI’s Trillion-Dollar Compute Web Blurs the Lines Between Procurement and Investment
New pacts with AWS, Oracle, Nvidia, CoreWeave and others lock in gigawatts of capacity as funding grows more complex and cash burn looms.
News
- Microsoft Launches 'World’s First AI Superfactory'
- OpenAI Pushes Back on NYT’s Demand for 20 Million ChatGPT Conversations
- SoftBank Sells Out of Nvidia as It Bets Big on AI
- UAE Pushes Ahead on 6G and Digital Dirham Adoption
- Meta’s Chief AI Mind Yann LeCun to Exit, Build Own AI Lab
- AMD’s Data Center Chip Revenue to Touch $1T by 2030, Overall Revenue to Grow by 35%
[Image source: Chetan Jha/MITSMR Middle East]
OpenAI has engineered large deals with Oracle, Nvidia, AMD, Broadcom, CoreWeave, and Amazon, with public and reported pledges topping $1 trillion over five years. The aim is to secure compute resources for its AI services, with approximately 30 gigawatts of planned capacity outlined across key suppliers.
Although deepening its access to high-performance infrastructure, the agreements commit OpenAI to significant spending, including approximately $300 billion with Oracle, up to $100 billion with Nvidia, roughly $38 billion with Amazon, and up to $22.4 billion with CoreWeave.
Multiple reports estimate the company’s 2025 annualized revenue at around $12–13 billion and its cash burn at nearly $8 billion, with approximately 70% of revenue coming from ChatGPT Plus and just over 5% of users paying. To manage its capital needs, the company is developing a five-year plan that includes new fundraising, debt, and additional revenue streams, according to the Financial Times.
Betting on Multiple Revenue Streams
To offset rising costs, OpenAI is experimenting with several revenue models. One is a potential advertising business, though executives maintain there are “no active plans” to launch ads. The company has, however, quietly hired talent from Google and Meta, suggesting a longer-term interest in monetizing its large user base.
It is also building shopping integrations with Walmart, Etsy, and Shopify, allowing users to make purchases directly through ChatGPT. While these partnerships could diversify revenue, consumer trust remains limited, and surveys indicate many users are hesitant to rely on AI for shopping decisions.
OpenAI is exploring a personal AI companion device with former Apple design chief Jony Ive, with some reports describing a screenless, smartphone-sized form factor. CEO Sam Altman has called it “a completely new kind of device,” designed to complement smartphones rather than replace them.
Beyond consumer products, OpenAI is building out its Stargate compute infrastructure with partners such as SoftBank, Oracle, Nvidia, and Cisco. The first Abilene, Texas site is guided at roughly 1.2 GW. A second facility, Stargate UAE, is slated to operate within Abu Dhabi’s 5-gigawatt AI campus.
Interestingly, the company is also expanding into the public sector. Its new ChatGPT Gov platform offers a secure version of ChatGPT for US federal agencies, delivered via Azure OpenAI Service on Microsoft’s commercial or Government clouds rather than OpenAI’s own servers. Likewise, it’s expanding its footprint in defense and national security, pursuing contracts and collaborations with partners such as Anduril — mirroring Anthropic’s work with Palantir.
Caution Over Overheating
Critics are increasingly voicing concerns about a potential AI bubble. One X user, HedgieMarkets, described OpenAI’s strategy as overextension, writing that “OpenAI needs trillions for infrastructure while burning $115 billion through 2029 on $5 billion annual revenue.”
The company’s $500 billion valuation, despite never posting a profit, has added to the skepticism. Even OpenAI’s chairman, Bret Taylor, has acknowledged the risks, warning that “we’re in a bubble and a lot of people will lose a lot of money.”
Jukan Lee, a South Korea-based technology analyst, noted on X that OpenAI’s trillion-dollar expansion “is not just about buying computers but also a carefully designed financial experiment.” According to Lee, OpenAI’s large supply deals with AMD and Nvidia effectively turn hardware purchases into complex investment instruments.
Under its agreement with AMD, OpenAI is not only purchasing GPUs worth about $90 billion but also securing the right to buy millions of AMD shares at discounted prices. If AMD’s stock rises, OpenAI could sell those shares to fund future GPU purchases, using market gains to offset hardware costs.
Analysts describe this as a self-reinforcing financial loop that ties AMD’s growth to OpenAI’s success.
The Nvidia partnership follows a similar pattern, though structured differently. Nvidia “intends to invest up to $100 billion” tied to a 10-GW rollout. While this boosts Nvidia’s revenue, some analysts refer to the arrangement as “circular revenue,” as the capital ultimately circulates between the same parties.
Goldman Sachs has warned that these intertwined structures depend heavily on external capital. OpenAI’s internal cash flow covers only a small share of total spending. Analysts estimate that a large majority of OpenAI’s financing relies on external capital. In short, the company’s vast AI ecosystem relies on continued support from investors and partners.
The market reaction has been dramatic. Oracle’s market value jumped by about $244 billion, and AMD’s by $63 billion, following news of their respective deals with OpenAI, but analysts caution the enthusiasm may not last.
Moody’s recently raised concerns about Oracle’s growing reliance on OpenAI-related contracts, while others note that Nvidia’s dependence on AI startups and sovereign wealth funds adds volatility to its earnings.
Growth at Any Cost
Despite the warnings, in a recent appearance on Brad Gerstner’s “Bg2 Pod” podcast, Altman dismissed fears about overextension, saying critics were exaggerating the scale of OpenAI’s infrastructure investments.
“I think there’s a lot of people who talk with a lot of breathless concern about our compute stuff or whatever that would be thrilled to buy shares,” he said. “We could sell your shares or anybody else’s to some of the people who are making the most noise on X about this very quickly.”
He added that the company expects strong growth ahead. “We do plan for revenue to grow steeply. We are taking a forward bet that it’s going to continue to grow.”
At the same time, speculation surfaced recently that OpenAI was preparing for a potential initial public offering (IPO) that could value the company at up to $1 trillion, one of the largest in history. OpenAI Chief Financial Officer Sarah Friar refuted the rumors, saying, “IPO is not on the cards right now… I don’t want to get wrapped around an IPO axle.”
Analysts said OpenAI’s performance is now deeply intertwined with the broader US technology market. As the Financial Times cautioned, any major disruption in OpenAI’s operations could ripple across the AI-driven economy.
Separately, as part of a broader restructuring, OpenAI recently completed a recapitalization that simplifies its corporate structure and reaffirms nonprofit control over its for-profit arm.
