Last September, Nvidia (NASDAQ:NVDA) and OpenAI lit up the AI world with the announcement of a letter of intent for a “landmark strategic partnership” to deploy at least 10 gigawatts of Nvidia systems, backed by up to $100 billion of progressive investment as capacity came online.

The agreement, though, was explicitly non‑binding and was dependent on infrastructure milestones such as data center and power build‑out. It also sat alongside a broader web of OpenAI arrangements and talks that, across cloud partners, chipmakers, and financial investors, have been reported to total approximately $1.4 trillion in potential commitments over multiple years.

However, The Wall Street Journal yesterday reported that Nvidia’s $100 billion plan had stalled amid internal doubts about the size and structure of the transaction and questions about OpenAI’s business discipline and competitive risks. Considering the interconnectedness of many of OpenAI’s agreements — and the criticism that Nvidia engages in a form of circular financing with its own deals — is this the loose thread that, once pulled, unravels the entire AI boom?

OpenAI’s position today is anchored in a dense network of partners and funding discussions that go far beyond any single Nvidia deal. Among the most prominent:

Microsoft (NASDAQ:MSFT) Azure: An incremental $250 billion commitment from OpenAI to purchase Azure cloud services for AI training and inference over several years.

Amazon (NASDAQ:AMZN) AWS: A seven-year $38 billion cloud computing partnership providing OpenAI access to AWS infrastructure (including Nvidia-equipped servers) for demanding AI workloads.

Oracle (NYSE:ORCL): A five-year, $300 billion agreement for cloud computing capacity and infrastructure that ties into the $500 billion Stargate Project to support OpenAI’s next-generation models.

CoreWeave (NASDAQ:CRWV): A $22 billion agreement for data center and GPU access — often Nvidia-powered — supporting OpenAI’s cloud needs.

Advanced Micro Devices (NASDAQ:AMD): A multi-year deal to supply around 6 GW of GPU capacity, potentially valued up to $300 billion. OpenAI would also receive warrants for a 10% stake in AMD if  certain targets are met.

On top of those strategic partners, OpenAI is also reportedly seeking up to $100 billion in fresh funding at a valuation around $830 billion, with potential contributions from Nvidia, Microsoft, Amazon, SoftBank, and Middle Eastern sovereign wealth funds.

In that context, the “up to $100 billion” Nvidia letter of intent was a cornerstone, but not OpenAI’s entire foundation.

Many of Nvidia’s deals have been criticized as a form of vendor financing. For example, Nvidia would invest huge sums in OpenAI, which then commits to lease or buy vast quantities of Nvidia chips. As Nvidia is also a major backer of CoreWeave — which buys Nvidia GPUs — it builds out data centers and then sells capacity to OpenAI and others.

This loop has led some observers to worry that revenues are being propped up by circularly financed demand rather than independent, sustainable economics. Nvidia rejects that characterization, emphasizing long‑standing commercial relationships and real, growing end‑user usage of AI services.

At first glance, the stalled $100 billion plan looks exactly like the scenario critics feared: if Nvidia pulls back its investment, perhaps OpenAI scales back its demand for CoreWeave’s capacity, which then reduces CoreWeave’s need for Nvidia chips. Rejecting an OpenAI investment means Nvidia just knocked over the first domino in the AI boom.

Several facts argue strongly against this view. First, the megadeal was non‑binding from the start. It was a letter of intent, not a signed obligation, that was contingent on deployment of 10 gigawatts of systems and associated infrastructure. OpenAI is also already in talks to raise up to $100 billion, with a significant portion expected from Microsoft, Amazon, Nvidia (at a reduced scale), SoftBank, and sovereign investors. The Nvidia shift changes the mix and terms, but not the existence of deep, competing pools of capital.

Demand for AI compute remains intense and megaprojects can be phased in. Multi‑gigawatt projects like the ones OpenAI discussed with Nvidia and data center partners are modular, so if one investor balks at funding the entire program at once, they can be built out in stages tied to utilization and revenue milestones instead of cancelled outright.

The Nvidia–OpenAI megadeal may end up smaller than advertised, or even be replaced by a patchwork of investments from multiple players, but that does not mean the AI gold rush is ending. If anything, it underscores that the ecosystem is broad, that OpenAI has many potential backers, and that partners will increasingly demand clearer paths to returns as the sector matures.

The boom in AI infrastructure and applications remains intact; what is changing is the willingness of even the biggest beneficiaries, like Nvidia, to keep writing blank checks. That looks less like the first domino falling and more like the AI era moving from exuberant promises into a more sober, long‑term growth phase.

For more than a decade, the investing advice aimed at everyday Americans followed a familiar script: automate everything, keep costs low, and don’t touch a thing. And increasingly, investors are realizing that being completely hands-off also means being completely disengaged.

That realization hits like a lightning bolt when you realize not just how much better your returns could be, but that there are amazing offers like one app where new self-directed investing accounts funded with as little as $50 can receive stock worth up to $1,000.

Take back your investing and start earning real returns, your way.