Nvidia Corp. Chief Executive Officer Jensen Huang spoke to media members after the company’s “Japan AI Ecosystem” reception on Thursday, July 16, 2026, in Tokyo, Japan.
Kiyoshi Ota | Bloomberg | Getty Images
Nvidia’s A huge head start in artificial intelligence has helped the chipmaker become the world’s most valuable company. Almost four years after the generative AI boom began, the competitors are: advanced micro device and google It chipped away at Nvidia’s technological edge and encouraged the company to tap into its other great asset: capital.
Following last week’s deal with Wall Street firms to seek $500 billion worth of funding for Nvidia’s graphics processing equipment, Nvidia announced Monday that it will provide up to $105 billion for its massive OpenAI data center in Ohio, providing a backstop of sorts in case ChatGPT’s developers see an upturn in their fortunes.
Nvidia’s strategy includes fostering the AI boom by any means necessary, recognizing that while the demand for critical infrastructure is seemingly insatiable, a small number of companies – hyperscalers – account for the vast majority of purchases. Nvidia, whose quarterly free cash flow has increased 18x over the past three years to $48.5 billion in the most recent period, is using the strength of its balance sheet and credit rating to avoid a dramatic slowdown despite 12 consecutive quarters of more than 55% revenue growth.
“They’re still ahead, but they’re very paranoid about not falling behind,” said Ram Bhalla, associate professor of AI and analytics at Santa Clara University’s Leavey School of Business.
Nvidia declined to comment.
Kantar analysts in a note to clients on Monday dismissed concerns that Nvidia is effectively buying revenue through its financial strategy. Both companies reiterated their Buy rating and said the agreement is a “clear sign that the current AI investment cycle is long and durable.”
“We view this as non-cyclical and as driving future AI builds while creating a further moat of competition that allows NVDA to remain an AI leader,” the analysts wrote.
Nvidia is swimming in money. The company’s cash generation is so great that in May it announced an increase in its quarterly dividend from 1 penny to 25 cents a share and announced a new $80 billion share buyback plan. The company pledged to return approximately 50% of its free cash flow to shareholders this year.
One way the company is putting its pile of cash to good use is through equity investments in companies across the AI ecosystem, including some of the companies spending big on Nvidia’s chips and systems, such as Model Developers and NeoCloud. Nvidia had $30.2 billion in marketable stock as of the most recent quarter, up from $12.9 billion a year earlier.
Nvidia invested $30 billion in OpenAI in February. OpenAI relies on the training capabilities of Vera Rubin, the chip giant’s cutting-edge system. Monday’s deal included a $1.5 billion investment in SB Energy, a SoftBank affiliate that is building and managing a data center at the PORTS-Pike Technology Campus in Pike County, Ohio, through a 20-year lease to OpenAI.
In addition to its investment in SB Energy, Nvidia said it will finance the development of about 4 gigawatts at the Ohio site for leases and a portion of the power and “certain residual value guarantees” as the data center opens between 2028 and 2030.
Expanding access
Nvidia CEO Jensen Huang acknowledged the importance of the company’s financial strength in a post to X about the deal.
“Frontier AI Labs has extraordinary demand for training and inference computation, much of which is growing faster than our balance sheet and long-term credit profile can support,” Huang wrote. “They may have strong customer demand and fast-growing revenues, but they still lack the decades-long infrastructure contracts and investment-grade financing capacity needed to secure their own AI factory infrastructure.”
A week ago, Huang was on the CNBC set, surrounded by six of Wall Street’s biggest financiers, announcing the arrival of NVIDIA graphics processing units as a new asset class. Signed a memorandum of understanding with companies including goldman sachsApollo Global Management, Blackstone, and BlackRock’s Huang suggested that the next phase of AI construction will be funded in part by third-party backers who can start investing in GPUs like they would in real estate.
“These are now income-producing assets,” Huang told CNBC. “It is highly productive, long-lived, fungible, and flexible.”
Huang’s dedication to the system will be key to obtaining financing for future borrowers, and NVIDIA will have the option to backstop 25% of all loans. This goes beyond just Google and AMD. cerebrum.
In the second quarter, Google began recognizing revenue from sales of TPU systems, contributing to 82% growth in its cloud division. Meanwhile, AMD is reporting more than 100% growth in its data center business, and its first rack-scale system, Helios, is scheduled to ship later this year.
Paul Meeks, head of technology research at Freedom Capital Markets, said increased competition is hurting Nvidia’s ability to generate “exorbitant profits” and motivating the company to diversify its strategy.
“Part of their idea is to expand our reach,” Meeks said. “We can’t ride one horse: the GPU.”
AI bulls say Nvidia is simply responding to demand and point out that today’s market shortfall is on the capacity side. There are many numbers to support this, human announced to investors over the weekend that its annual revenue run rate reached $65 billion in July, a sevenfold increase from a year earlier. OpenAI The run rate recently reached $40 billion.
Matthew Begari, director of research at Clearwater Analytics, said in an email that given market trends, “the narrative of a roundabout ‘house in the sand’ structure for the AI industry appears to be somewhat misplaced.”
“One day we may exceed our production capacity,” he wrote. “But that day is not today.”
— CNBC’s Samantha Subin and Jonathan Bunyan contributed to this report
WATCH: AI is not a new asset class, it’s an entire market, says Clockwise Capital CIO
