TradingKey - After raking in massive profits from selling AI chips, Nvidia (NVDA) is now transforming from a hardware supplier into an architect of the AI ecosystem through aggressive infrastructure investments. The company’s latest strategic partnership with OpenAI, involving a potential $100 billion investment, deepens its ties with the world’s most valuable tech startup.
On Monday, September 22, Nvidia and OpenAI announced a letter of intent to form a strategic alliance. Under the agreement, OpenAI will use Nvidia’s systems to build and deploy at least 10 gigawatts (GW) of AI data centers, powered by millions of Nvidia AI processors to train next-generation AI models.
This capacity is equivalent to:
Nvidia CEO Jensen Huang said this scale matches Nvidia’s total GPU shipments for the entire year, which is double last year’s volume.
At the same time, Nvidia will invest up to $100 billion in OpenAI in exchange for equity — making it Nvidia’s largest-ever external investment and one of the biggest private-company investments in history.
Currently, Nvidia’s market cap stands at approximately $4.5 trillion, while OpenAI’s valuation on the private trading platform Forge Global is $324 billion, far ahead of second-place Anthropic ($178 billion).
The announcement ignited investor enthusiasm, with Nvidia shares turning positive after an early dip, closing up 3.93% at a new all-time high. Year-to-date, Nvidia is up about 37%, outpacing the Nasdaq’s 18% gain.
OpenAI CEO Sam Altman said the deal will unlock new breakthroughs in AI by securing essential computing power.
“Everything starts with compute,” he said.
Bernstein analyst Stacy Rasgon estimates that one gigawatt of data center capacity could involve tens of billions of dollars’ worth of Nvidia products. The Financial Times noted the deal could generate hundreds of billions in revenue for Nvidia over time.
Although Nvidia continues to report strong revenue growth, it remains heavily reliant on a few major clients — Microsoft, Amazon, Google, and Meta — all of whom are developing in-house alternatives to reduce dependence on Nvidia. This customer concentration risk has become increasingly evident.
Adding OpenAI as a core client provides Nvidia with a powerful new anchor tenant — diversifying its customer base and reinforcing long-term demand.
MIT professor Michael Cusumano described the move as a kind of “wash”:
“Nvidia is investing $100bn in OpenAI stock and OpenAI is saying they are going to buy $100bn or more of Nvidia chips.”
OpenAI will use Nvidia’s capital to build data centers, while Nvidia earns back-end hardware sales — creating a self-reinforcing cycle. While such circular arrangements are common in AI infrastructure deals, they have also sparked debate over whether AI demand is truly sustainable or artificially inflated.
Huang called the partnership a “next leap” in the AI era.
While a $100 billion investment may seem staggering, unlike OpenAI’s $300 billion commitment to Oracle — whose delivery capability remains uncertain — Nvidia has the financial muscle to deliver. Analysts project Nvidia’s free cash flow for the current fiscal year to be on par with $100 billion.
S&P Global Visible Alpha said this is a fully feasible investment, citing that Nvidia has extremely strong cash flow, a solid balance sheet, and a proven track record of sustained growth.
Some analysts view this as part of Nvidia’s broader strategy to diversify not just geographically (in supply chains), but strategically across the AI stack — positioning itself as a foundational infrastructure player.
Nvidia’s recent moves underscore this shift:
PitchBook analysts commented that Nvidia is tightening its grip on the AI stack and cementing its role as an indispensable enabler of the industry — and the OpenAI deal is the latest proof.