OpenAI executives are fielding a number of questions about how the company plans to pay for the $1.4 trillion worth of data center construction and usage contracts it accrued this year, given its rapidly increasing revenue and $20 billion annual utilization rate, CEO Sam Altman said Thursday in a post on X.
Altman’s comments were in response to comments from OpenAI CFO Sarah Friar, who quickly retracted her response. Speaking at a Wall Street Journal event on Wednesday, Fryer said he wants the U.S. government to “backstop” its own infrastructure financing. This makes financing cheaper for the company and ensures it always has the latest and greatest chips, she explained.
Backstop loans are guaranteed by the government, so if a company defaults, taxpayers will pick up the slack. Lenders tend to reward such low-risk loans with better terms.
Fryer said that while computing-constrained OpenAI’s need to use older chips makes funding options more affordable, the company’s goal is always to have cutting-edge models on the latest and greatest chips.
So how will this revolving door of chips be paid for? He said the company is looking for an “ecosystem” to support it, including banks, PE firms and even governments.
When asked what she would like the government to do, she replied: “…a backstop, or a guarantee that makes a loan possible. This not only significantly lowers the cost of financing, but also increases the value of the loan, thereby increasing the amount of debt that can be assumed in addition to equity.”
She also hinted that such talks were already underway, particularly in the United States, saying, “I think we’re seeing that. The U.S. government in particular has been incredibly proactive and really understands that AI is almost a national strategic asset.”
tech crunch event
san francisco
|
October 13-15, 2026
Fryer quickly retracted her comments after the Wall Street Journal published a clip of her discussing her desire for a federal backstop, and many X users with large followings scoffed at the idea.
“I would like to clarify my comments today: OpenAI is not seeking a government backstop for our infrastructure efforts. I used the word ‘backstop’ to obscure the point,” she posted on LinkedIn.
On Thursday, President Trump’s AI czar, David Sachs, also weighed in. Sachs, himself a major Silicon Valley venture capitalist, wrote in X that the US has no plans to bail out any AI companies.
“There will be no federal bailout for AI. There are at least five major frontier model companies in the US. If one fails, others will take their place,” he wrote, adding that what the government wants is “to make permitting and generation easier.” Although he did not name her, he also allowed Friar to “clarify” her position.
In response, Altman wrote a lengthy post about X, echoing Sacks.
“We do not have and do not want government guarantees for OpenAI data centers. We believe governments should not pick winners and losers, and taxpayers should not bail out companies that make bad business decisions or lose money in the market,” he wrote.
He also revealed that while backstop financing has been discussed, his company has not.
“One area where we have been discussing loan guarantees is to help build semiconductor factories in the United States, and we and other companies have responded to the government’s call and are happy to help (although we have not formally applied).”
It’s hard to blame Friar for coming up with this idea. Even though, as Sacks writes in her article, the idea of asking taxpayers for a bailout is “ridiculous,” she is right that such a guarantee would make the job of raising money easier.
Now that she’s hearing a loud public “no” from the people she wants for her idea, she and OpenAI CEO Sam Altman can expect to face even more questions about how they plan to pay for their $1 trillion expansion.
In fact, Altman appears to be preparing for just such a thing.
“We expect to end this year with an annual revenue run rate of more than $20 billion and grow to hundreds of billions of dollars by 2030. We are looking at commitments of approximately $1.4 trillion over the next eight years,” he wrote, adding that the company was pleased with its “outlook” for enterprise products, new consumer devices, robotics, and more, among others.
