I’ve been watching Nvidia’s investment strategy unfold over the past few months, and honestly, I’m not sure if I’m witnessing brilliant foresight or the tech industry’s most elaborate money laundering scheme.
The numbers are staggering. Nvidia has committed over $40 billion to equity investments in AI companies just in these early months of 2026. That’s not a rounding error. That’s a fundamental shift in how one of the world’s most powerful chipmakers is playing the game.
Let me break down what’s happening here. The headline number gets a lot of attention because it’s anchored to one massive bet: a $30 billion investment in OpenAI. That’s nearly three-quarters of their entire stated AI investment allocation going to a single company. But the story gets more interesting (and more troubling, depending on your perspective) when you look at the rest of the portfolio.
The Pattern Nobody Wants to Admit
Beyond OpenAI, Nvidia has been spreading its capital across two main categories. There’s the venture side, where they participated in around 67 investment rounds in 2025, and they’re already in two dozen more rounds this year in private startups. Then there’s the publicly traded companies, where they’ve announced seven separate multi-billion dollar deals. The most recent ones caught my eye: $3.2 billion into Corning and $2.1 billion into IREN, a data center operator.
Here’s where it gets uncomfortable. Many of these companies are Nvidia’s customers. Some are potential customers. Some are both.
Matthew Bryson from Wedbush Securities called it what everyone’s thinking but nobody wants to say out loud: these deals fall “squarely into the circular investment theme.” The concern isn’t paranoia. When you’re investing $30 billion into the company that’s going to buy your chips, and then investing billions more into the infrastructure companies that support that ecosystem, you’re creating something that looks a lot less like venture capitalism and a lot more like a closed loop where money just bounces between the same players.
But here’s the thing that makes me pause before dismissing it entirely: Bryson also noted that if these investments succeed, they could genuinely help Nvidia build a “competitive moat.” And that’s the billion-dollar question everyone should be asking.
The Moat Argument, and Why It Matters
Think about what Nvidia is actually doing when you step back. They’re not just buying chips to companies anymore. They’re becoming a stakeholder in the entire value chain of artificial intelligence infrastructure.
If you invest in OpenAI and they dominate the LLM space, guess whose chips they’ll buy by the millions. If you invest in Corning and they become the dominant supplier of specialty glass for data centers, and you invest in IREN and they become a major data center operator, you’ve created a situation where the entire ecosystem is partially owned by you.
It’s not exactly a moat in the traditional sense. It’s more like owning the castle, the surrounding land, and a chunk of the royal family’s treasury all at once.
From a technical standpoint, this is fascinating. AI companies are GPU-constrained. Data center operators need specialized equipment. Glass manufacturers need specific partnerships. By being a meaningful equity holder in multiple parts of this value chain, Nvidia gets board seats, insights into roadmaps, and preferential treatment on allocations. When supply is limited and competition is fierce, having that kind of inside information and structural advantage is worth far more than the capital deployed.
The counter-argument is the one critics keep making: this is just Nvidia finding creative ways to ensure demand for their products stays artificially high. It’s like owning the gas station and the car dealership. Sure, you make money on both, but you’re not really competing. You’re just taking different cuts of the same revenue stream.
The Venture Reality Check
What fascinates me about the venture side of this equation is how different it looks from the mega-bets. Nvidia participated in 67 venture deals in 2025. That’s a diversified approach that suggests someone internally is thinking about ecosystem building rather than just returning a check to friendly companies.
Most of these startups won’t be customers. Many will fail. Some will compete with Nvidia’s own interests. That’s real venture investing. That’s risk capital.
The publicized mega-deals though? Those are different animals. A $30 billion check to OpenAI isn’t venture investing in the traditional sense. It’s strategic positioning. It’s ensuring Nvidia isn’t left out of the most valuable AI company in the world while making sure that company has the capital to buy even more of their chips.
I don’t think this is necessarily wrong or corrupt. But I do think it’s important to be clear about what it actually is, because the venture framing obscures what’s really happening.
What This Means for Developers and Startups
Here’s what I’d be thinking about if I were building a startup right now: Nvidia’s ability to deploy $40 billion in capital means they can pick winners. And more importantly, it means they can create or sustain positions in companies that might otherwise struggle.
If you’re building an AI startup and Nvidia invests in you, that’s validation. That’s also a relationship with the company that controls the chips you probably need. That’s powerful. It’s also worth thinking carefully about what that relationship looks like, what expectations come with it, and whether you’re comfortable with a major customer also being a significant shareholder.
For larger companies in this space, Nvidia’s investment strategy is a signal. They’re not just going to be a chip vendor. They’re going to be embedded in your company’s governance, strategy, and long-term planning. Whether that’s good or bad depends entirely on your perspective and your specific situation.
The real question isn’t whether Nvidia is doing something wrong. The question is whether the market will eventually reach a point where this kind of structural advantage becomes untenable. Will regulators care? Will other chip companies follow suit? Will the companies receiving these investments get tired of having Nvidia in their shareholder base?
Those answers will probably determine whether this strategy looks like visionary ecosystem building or circular deal-making that eventually collapses under its own weight.