CaliToday (20/10/2025): OpenAI, the trailblazing company behind ChatGPT, is on a trajectory to spend over $1.3 trillion by 2030 to secure the massive computing power it has promised, a new analysis from Citi warns. This staggering sum, however, faces a critical reality check: the company's projected revenue comes nowhere close to covering the cost, sparking fresh concerns on Wall Street about a rapidly inflating AI bubble.
 |
| OpenAI CEO Sam Altman speaks during Snowflake Summit 2025 in June 2025 in San Francisco. (Justin Sullivan/Getty Images) |
The research note from Citi analysts highlights the astronomical capital expenditures required for the next phase of the artificial intelligence revolution. OpenAI has entered into massive, multi-year partnerships with chipmakers Nvidia (NVDA), Broadcom (AVGO), and Advanced Micro Devices (AMD) to build out its AI infrastructure.
The core of these deals is an ambitious promise to deploy 26 gigawatts worth of computing capacity.
To put that figure in perspective, 26 gigawatts is nearly the amount of power required to provide electricity to the entire state of New York during peak summer demand.
A $1.3 Trillion Calculation
The path to this trillion-dollar figure is based on a sobering cost analysis. Citi analysts estimate that it currently takes approximately $50 billion in total spending to bring just one gigawatt of AI compute capacity online. This cost includes not only the advanced chips but also the massive supporting energy infrastructure and data center construction.
Based on that assumption, Citi analyst Chris Danely warned clients this week that to reach its 26-gigawatt goal, OpenAI's capital expenditures would hit $1.3 trillion by 2030.
Disturbingly, even that figure may be conservative. OpenAI CEO Sam Altman has reportedly floated far bolder promises internally. The Information reported in late September that the executive has suggested the company is ultimately looking to deploy 250 gigawatts of computing capacity by 2033.
If that 250-gigawatt figure is accurate, the implied cost, based on Citi's model, would be a mind-boggling $12.5 trillion.
The Wall Street "Disconnect" and Bubble Fears
The central problem for OpenAI, and by extension the entire AI-driven market, is the yawning gap between costs and revenue.
"But there's no guarantee that OpenAI will have the capital to support the costs required to achieve its goals," the Citi note stated.
While OpenAI's costs are projected to soar past the trillion-dollar mark, Citi estimates the company's revenue will climb to only a fraction of that figure—$163 billion—by 2030.
This massive "disconnect" has added significant fuel to Wall Street concerns over a stock market bubble. Global stocks have soared to new records this year, largely propelled by investor optimism and speculative fervor surrounding artificial intelligence. The idea that the market leader's business model could be based on expenditures that outpace revenue by nearly 8-to-1 has unnerved many.
A "Tangled Web" of AI Bets
This $1.3 trillion commitment is not the beginning of OpenAI's spending. It's an acceleration. The company has already made colossal commitments to the global AI build-out:
Project Stargate: A $300 billion deal announced in September with Oracle (ORCL) as part of a 10-gigawatt U.S. AI infrastructure project.
Global Expansion: Additional "Stargate" projects have been unveiled abroad in partnership with Nvidia in the United Arab Emirates and Norway.
Data Center Capacity: A $22 billion commitment to purchase capacity from CoreWeave (CRWV), an AI data center provider that is itself backed by Nvidia.
This tangled web of investments, where major players are simultaneously partners, customers, and investors in each other, has led to concerns that AI demand could be artificially overstated.
The stakes, according* to* Bernstein analyst Stacy Rasgon, could not be higher. In an October 6 note, Rasgon wrote that Sam Altman "has the power to crash the global economy for a decade or take us all to the promised land, and right now we don't know which is in the cards."
The Physical and Financial Hurdles
Beyond the immediate funding concerns, it remains unclear whether the U.S. power infrastructure can physically scale up in time to meet this explosive energy demand. If the grid fails to keep pace, these multi-trillion-dollar AI projects would be unable to come online, preventing OpenAI from ever cashing in on its spending.
If, however, OpenAI succeeds in finding the capital and the power, the chipmakers are positioned for a windfall of historic proportions. Bank of America analyst Vivek Arya estimates Nvidia alone could see as much as $500 billion in revenue from OpenAI if the total deal amount is fulfilled. Rasgon, at Bernstein, estimated Broadcom could see over $100 billion in revenue from its own deal with the ChatGPT developer.
CaliToday.Net