OpenAI Just Raised $110 Billion and the AI Arms Race Has Never Looked More Insane

The Biggest Check in Tech History

Let that number sink in: $110 billion. That's what Amazon, Nvidia, and SoftBank collectively poured into OpenAI in a single funding round announced on February 27, valuing the company at $730 billion pre-money (and roughly $840 billion post-money). To put that in perspective, this is the largest private funding round in the history of technology, in the history of venture capital, in the history of anything that isn't a sovereign government issuing bonds. OpenAI is now worth more than most countries' GDPs.

The breakdown: Amazon led with a staggering $50 billion commitment, Nvidia contributed $30 billion, and SoftBank matched Nvidia at $30 billion. Amazon's investment starts with an initial $15 billion, with the remaining $35 billion contingent on "certain conditions being met." In October, OpenAI was valued at $500 billion during a secondary financing. Five months later, that number jumped by $230 billion. Even by the standards of Silicon Valley's most inflated era, this is something entirely new.

Amazon Enters the Chat

The most fascinating part of this deal isn't the size; it's the partnerships baked into it. Amazon and OpenAI announced a strategic partnership that positions AWS as the exclusive third-party cloud distribution provider for OpenAI Frontier, the company's enterprise platform for building, deploying, and managing AI agents. OpenAI committed to consuming at least 2 gigawatts of Amazon's Trainium AI chips and is expanding a previous $38 billion AWS agreement by an additional $100 billion over eight years.

That's right: OpenAI is spending $100 billion on Amazon's cloud infrastructure over the next eight years. Analysts estimate that works out to roughly $17 billion a year in AWS revenue if spending is spread evenly, which would represent about 11% of AWS's expected 2026 revenue. For Amazon, this isn't just an investment; it's a customer acquisition play of historic proportions.

What About Microsoft?

The obvious question: what happens to Microsoft, OpenAI's closest partner since 2019? Microsoft and OpenAI moved quickly to reassure the market, publishing a joint statement emphasizing that nothing has changed. Azure remains the exclusive cloud provider for stateless OpenAI APIs, including ChatGPT and the consumer products. Microsoft's revenue sharing terms are unchanged, and its IP licensing arrangement continues as before.

Microsoft went so far as to say it was "excited to see what they build together," which is either a remarkable display of corporate graciousness or the kind of carefully worded diplomatic statement that masks deep strategic concern. The reality is more nuanced: Microsoft's Azure handles OpenAI's existing API business, while Amazon's AWS gets the new enterprise agent platform. OpenAI has effectively split its cloud business between the two biggest providers, creating competition for its own spending. Whether that creates leverage for OpenAI or awkwardness for everyone remains to be seen.

The Nvidia Connection

Nvidia's $30 billion investment comes with its own infrastructure deal. OpenAI gets 3 gigawatts of dedicated inference capacity and 2 gigawatts of training on Nvidia's next-generation Vera Rubin systems. For Nvidia, this is the ultimate circular deal: invest $30 billion in your biggest customer, who then spends that money (and much more) buying your chips. It's a financing arrangement that ensures demand for Nvidia's most advanced hardware while giving OpenAI guaranteed access to the GPUs it needs.

This circular financing pattern is becoming the defining feature of the AI industry. Chipmakers and cloud providers back AI startups who are also their largest customers. The money flows in a loop: investors put cash in, the startup spends it on chips and cloud, which generates revenue for the investors, which justifies higher valuations, which enables more fundraising. It works beautifully as long as the end demand for AI products justifies the spend. If it doesn't, the entire system unwinds.

The Burn Rate Problem

Here's where the optimism gets tested. OpenAI reported $13.1 billion in revenue for 2025 and projects $30 billion for 2026. Those are impressive growth numbers. But the company also expects to burn through a staggering $218 billion between 2026 and 2029, which is about $111 billion more than internal projections from just two quarters ago. The burn rate is expected to remain at roughly 57% through 2026 and 2027.

Inference costs alone are projected to hit $14.1 billion in 2026. Gross margins hover around 40%, constrained by the massive variable compute costs of running models at scale. OpenAI expects to post "stunning annual losses" through 2028 before turning "wildly profitable" in 2030, according to internal financial documents obtained by Fortune. The company currently has about 700 million weekly ChatGPT users, but only about 35 million (5%) pay for subscriptions.

The fundamental question hasn't changed: can OpenAI convert usage into enough revenue to justify spending hundreds of billions of dollars on infrastructure? The $110 billion raise buys time, but it doesn't answer the question.

What the Capital Market Is Saying

The broader capital system has clearly decided that Sam Altman's premise is correct: revolutionary technology requires unprecedented funding. The investors in this round aren't speculative venture funds; they're the three companies (Amazon, Nvidia, SoftBank) best positioned to benefit from AI infrastructure buildout regardless of which AI company ultimately wins. Amazon gets a massive new cloud customer. Nvidia gets guaranteed chip demand. SoftBank, which has been the most aggressive AI investor globally under Masayoshi Son, gets a stake in the company that arguably defines the industry.

But the concentration of power is striking. With Amazon, Nvidia, and SoftBank all holding significant stakes in OpenAI while simultaneously being suppliers, the conflicts of interest are stacked high. These are the same companies that sit on the other side of the table during every infrastructure negotiation. The FTC hasn't flagged this specific deal yet, but antitrust scrutiny of these interlocking AI investments is a growing concern in Washington.

The Competition Landscape

OpenAI's raise puts enormous pressure on its competitors. Anthropic, Google, and Meta all have to match this spending trajectory or risk falling behind on model capabilities. Anthropic raised its own significant round recently but is operating at a fraction of OpenAI's budget. Google has the advantage of internal infrastructure through Google Cloud, but even Google's AI spending is drawing investor concern. Meta is building AI largely on the open-source model, which requires less fundraising but also generates less direct revenue.

The irony is that OpenAI's record raise happened the same week that Trump's administration banned Anthropic from Pentagon contracts, handing OpenAI the inside track on military AI spending. Whether by design or coincidence, OpenAI's position as the U.S. government's preferred AI partner is strengthening, and that political alignment may matter as much as the technology.

What to Watch

The March 4 Apple event is the next big moment for consumer AI. Nvidia's GTC conference starts March 16, where Jensen Huang has promised "several new chips the world has never seen before." But the real test for OpenAI is quieter: can the company show a credible path to reducing that 57% burn rate while maintaining the research velocity that justifies a $730 billion valuation? The $110 billion buys roughly 2-3 years of runway at current burn rates. If AI revenues don't scale dramatically by 2028, this becomes the most expensive bet in corporate history. And if they do scale, everyone who thought this was insane will look foolish in hindsight. That's the nature of transformative technology: the correct amount of investment always looks unreasonable until it doesn't.