Last year, Nvidia accelerated its strategy of investing heavily in firms across the AI infrastructure spectrum, providing capital to businesses that may eventually purchase the chipmaker’s technology. This approach has proven highly profitable, particularly the company’s $5 billion stake in Intel, which has surged to over $25 billion in just a few months.

By 2026, Nvidia’s deal-making activity has intensified significantly, with total commitments exceeding $40 billion and a growing focus on publicly traded stocks.

Earlier this week, Nvidia announced a $2.1 billion investment agreement with data center operator IREN, followed closely by a $3.2 billion pact with Corning, a century-old glass manufacturer. Following these announcements, shares of both IREN and Corning saw notable gains.

Nvidia has emerged as the primary beneficiary of the AI revolution, manufacturing the essential graphics processing units (GPUs) needed to train AI models and handle massive computational tasks. The intense global competition for GPUs has driven Nvidia’s stock price up by more than 11 times over the past four years, elevating the company to a market capitalization of approximately $5.2 trillion and making it the world’s most valuable enterprise.

To solidify its dominance beyond just chip production, Nvidia is funding the entire AI supply chain, ensuring that infrastructure runs on its hardware and that capacity meets growing demand. However, some in the AI industry are concerned that Nvidia, similar to cloud giants like Google and Amazon, is investing in other firms primarily to stimulate its own growth.

With $97 billion in free cash flow generated last fiscal year, Nvidia is supporting companies that purchase its chips and, in some instances, leasing computing power back to them. Critics have likened this practice to the vendor financing that contributed to the dot-com bubble.

Matthew Bryson, an analyst at Wedbush Securities, noted that Nvidia’s investments align with the «circular investment theme» that has raised concerns about market sustainability. Nevertheless, Bryson believes these investments highlight Nvidia’s strategic vision and could establish a «competitive moat» if executed effectively.

An Nvidia spokesperson did not respond to requests for comment.

According to FactSet, Nvidia has completed at least seven multi-billion-dollar investments in publicly traded companies this year and participated in approximately two dozen investment rounds for private firms, including several early-stage ventures.

‘We don’t pick winners’

Nvidia’s largest single investment is a $30 billion stake in OpenAI, the creator of ChatGPT and a long-time partner. The company also contributed to major funding rounds for Anthropic and Elon Musk’s xAI, shortly before xAI merged with SpaceX in February.

«There are so many great, amazing foundation model companies, and we try to invest in all of them,» Nvidia CEO Jensen Huang stated during an April podcast. «We don’t pick winners. We need to support everyone.»

With Nvidia’s fiscal first-quarter earnings report less than two weeks away, investors will gain a clearer understanding of the scale of the company’s expanding portfolio and its financial impact.

During the previous fiscal year, Nvidia invested $17.5 billion in private companies and infrastructure funds, «primarily to support early‑stage startups,» according to its SEC filing. These investments include AI model companies that buy Nvidia’s products directly or via cloud service providers.

Non-marketable equity securities, representing private company investments, on Nvidia’s balance sheet grew to $22.25 billion by the end of January, up from $3.39 billion a year prior. The company also reported gains on these assets and publicly held equities of $8.92 billion, up from $1.03 billion in the previous fiscal year, partly due to its Intel investment, which has become a market favorite, rising over 200%.

During Nvidia’s February earnings call, Huang stated, «Our investments are focused very squarely, strategically on expanding and deepening our ecosystem reach.»

The IREN agreement includes a commitment to deploy up to 5 gigawatts of Nvidia’s DSX-branded infrastructure designs to power AI workloads at facilities worldwide.

Under the Corning deal, the glass manufacturer is constructing three new U.S. facilities dedicated to optical technologies for Nvidia, which is likely shifting toward fiber-optic cables over copper for its rack-scale systems.

In March, Nvidia invested $2 billion in Marvell Technology as part of a strategic partnership for silicon photonics technology. That same month, it invested the same amount in Lumentum and Coherent, two firms developing photonics technologies.

Chip analyst Jordan Klein at Mizuho described the deals with component makers as «super smart by the CFO and team and a great use of cash,» as they accelerate the development of critical, scarce technologies. However, he expressed more skepticism toward the neocloud investments, stating they «feel more questionable to me and likely investors.»

«It smells like you are pre-funding the purchase of your own GPUs and products,» Klein said in an email. Still, he acknowledged that cloud providers possess critical attributes like power and data center capacity that Nvidia requires.

Ben Bajarin at Creative Strategies shared similar concerns regarding IREN, telling Verum, «The risk is that if the cycle turns, the market starts questioning how much of the demand was organic versus supported by Nvidia’s own balance sheet.»

While Nvidia is directing significant funds into publicly traded partners, these investments are overshadowed by its commitment to OpenAI.

Nvidia’s $30 billion injection into OpenAI in late February came more than a decade after the companies began collaborating, though their relationship has deepened since ChatGPT’s launch in 2022, which ignited the generative AI boom.

Nvidia’s initial investment in OpenAI was intended to be much larger. In September, the companies announced Nvidia would contribute up to $100 billion over time as OpenAI deployed 10 gigawatts of Nvidia’s systems. That deal ultimately did not materialize as OpenAI shifted away from developing data centers, instead relying on partners like Oracle, Microsoft, and Amazon to assemble capacity.

Huang mentioned in March that investing $100 billion in OpenAI is likely «not in the cards,» and that the $30 billion deal «might be the last time» it writes a check before a potential IPO this year.

WATCH: Nvidia’s AI supply chain empire: Here’s what you need to know