🔥 WHAT HAPPENED

Nvidia isn't just the pick-and-shovel seller of the AI gold rush anymore. It's become the venture capitalist writing the checks.

In the first five months of 2026, Nvidia has committed over $40 billion to AI equity investments, according to a CNBC exclusive published Friday. That's more money than most countries' GDP, and it's all been deployed in a single year so far.

The centerpiece? A $30 billion investment in OpenAI — a bet so large it makes most tech acquisitions look like pocket change.

But that's just the headline. Dig deeper and you'll find Nvidia has signed at least seven multi-billion-dollar deals with publicly traded companies this week alone, plus roughly two dozen private startup rounds tracked by FactSet. The company is spending like a nation-state building an empire.

🧠 WHY THIS MATTERS

Nvidia is quietly turning itself into the financial backbone of the entire AI industry. It's not just selling the shovels — it's financing the mines, the logistics, the refineries, and the end product.

Think about it from Jensen Huang's perspective. Nvidia generated $97 billion in free cash flow last fiscal year. That's a staggering pile of money, and the company needs to put it to work. The established playbook would be stock buybacks or dividends. Instead, Nvidia is placing strategic bets across the full AI stack — model companies, data center operators, cloud providers, component manufacturers, and optical networking firms.

The logic is simple: ensure every layer of the AI supply chain runs on Nvidia hardware. If you're an AI startup building on Nvidia chips, and the data centers are built with Nvidia-backed infrastructure, and the optical cables connecting everything are made in Nvidia-funded factories — you're not going to switch to a competitor.

Here's the real kicker: Nvidia's $5 billion bet on Intel is now worth over $25 billion. That's a 5x return in months. The company isn't just financing the ecosystem — it's getting filthy rich doing it.

📊 DEEP DIVE

Let's break down the deals that made up this insane week alone:

Nvidia committed $30 billion to OpenAI in a deal that was originally supposed to be significantly larger. Back in September, both companies said Nvidia would put in up to $100 billion over time. Huang walked that back in March, calling that figure "not in the cards" and hinting the $30 billion might be Nvidia's last check before a potential OpenAI IPO this year.

On May 6, Nvidia struck a deal with the 175-year-old glassmaker to invest up to $3.2 billion. Corning is building three brand-new US factories in North Carolina and Texas, dedicated entirely to optical technologies — fiber-optic cables and connectors — for Nvidia's rack-scale AI systems. The shift from copper to fiber is critical for AI infrastructure, and Nvidia is literally building the factories to make it happen.

The very next day, Nvidia forged an agreement with data center operator IREN for the right to invest up to $2.1 billion. IREN will deploy up to 5 gigawatts of Nvidia DSX-branded AI infrastructure globally. That's enough power to run multiple cities.

In March, Nvidia invested $2 billion in Marvell to collaborate on silicon photonics — a technology that uses light instead of electricity to move data between chips. It's bleeding-edge stuff that could define the next generation of AI hardware.

Nvidia put $2 billion into CoreWeave and another $2 billion into Nebius Group, both AI cloud companies. These "neoclouds" are essentially building data centers designed around Nvidia's technology stack, giving Nvidia's customers more places to run their workloads.

Also in March, Nvidia invested $2 billion each in Lumentum and Coherent — two more photonics companies working on the optical technologies that will connect AI systems at massive scale.

Add it up: that's roughly $47 billion in known deals already announced, and the year isn't even half over.

⚠️ THE CATCH

Analysts are raising eyebrows at Nvidia's rapid transformation from chipmaker to venture capitalist. The criticism boils down to one phrase: circular investments.

Here's how the circular argument works: Nvidia invests in a data center company. That data center company uses Nvidia's money to buy Nvidia's GPUs. The data center company then rents computing power to AI model companies — some of which Nvidia has also invested in. Those AI companies pay the data center company with money they got from Nvidia.

"Squarely into the circular investment theme," is how Wedbush analyst Matthew Bryson described it. Critics compare it to the vendor financing that inflated the dot-com bubble. When everyone's money is everyone else's money, what happens when the music stops?

Mizuho chip analyst Jordan Klein called the component-maker deals "super smart by the CFO and team" but was more skeptical about the neocloud investments. "It smells like you are pre-funding the purchase of your own GPUs and products," he said.

But there's a flip side. Bryson also noted these investments could create a "competitive moat" if Nvidia executes well. And given their track record, betting against Jensen Huang has been a historically bad idea.

🎯 WHAT HAPPENS NEXT

Nvidia's fiscal first-quarter earnings are due in less than two weeks. That's when shareholders will get a clearer picture of the portfolio's size and its actual impact on financials.

The company's non-marketable equity securities ballooned from $3.39 billion to $22.25 billion in a single year. Investment gains hit $8.92 billion, up from $1.03 billion — largely thanks to the Intel bet that went nuclear.

Meanwhile, Huang has suggested the OpenAI investment "might be the last time" Nvidia writes a check before OpenAI goes public. An OpenAI IPO this year would mark one of the biggest public listings in history, and Nvidia would be holding a $30 billion stake.

For the broader market, keep an eye on Samsung crossing the $1 trillion valuation mark, powered by AI chip demand. Samsung's chip profit jumped almost 50-fold, and the company warns that the memory shortage will worsen in 2027. That means Nvidia's infrastructure buildout is happening just in time — or maybe not fast enough.

🧩 BIGGER PICTURE

Nvidia is building something the tech world has never seen: a vertically integrated AI empire with a $5.2 trillion market cap, $97 billion in annual free cash flow, and tentacles in every company that matters. It's part chipmaker, part VC firm, part infrastructure builder, and part nation-state. Altimeter Capital CEO Brad Gerstner thinks it could become the world's first $10 trillion company.

The circular money argument is real. But it's also a feature, not a bug. Nvidia's strategy ensures that AI infrastructure grows on its terms, using its technology, powered by its chips. If you're competing with Nvidia's ecosystem, you're not just competing with a chip company — you're competing with a $40 billion war chest that touches every corner of the industry.

The question isn't whether Nvidia can afford to keep doing this. It's whether anyone can afford to stop them.