Nvidia's AI equity investments hit $99 billion in one year

A chipmaker becomes an AI financier

Nvidia’s role in the artificial intelligence boom is no longer limited to selling GPUs, networking gear and software stacks. The company’s equity investments tied to the AI ecosystem reached $99 billion as of July 26, up from about $7 billion a year earlier and roughly $2.2 billion two years earlier, according to fresh reporting published Friday [1]. That increase, more than tenfold in one year, reframes Nvidia as both the dominant supplier of AI compute and a major capital provider to the companies expected to consume it.

The scale matters because it changes how investors should read Nvidia’s growth. A company that once benefited primarily from customers’ capital expenditures is now helping to shape the balance sheets of those customers and partners. The reported $99 billion portfolio spans frontier model labs, AI infrastructure financiers, neocloud providers, photonics and optical companies, semiconductor partners and application-layer AI firms [1]. In other words, Nvidia is backing the demand chain around its own platforms.

The headline figure also lands at a moment when Nvidia is trying to broaden the market for accelerated computing. In recent days, the company has been linked to a nearly $13 billion deal for Hugging Face, a $3.5 billion convertible-bond investment in MediaTek, and fresh attention to its financing relationships with AI labs and cloud-compute specialists [3][4]. Together, those moves point to a strategy: keep Nvidia hardware and interconnects at the center of AI even as customers experiment with custom chips, open-source models and alternative infrastructure models.

What the $99 billion figure includes

The reported total covers equity investments in AI-related companies and sits far above Nvidia’s level a year ago [1]. Market Chatter reporting carried by Yahoo Finance also said Nvidia’s equity investments had risen to $99 billion as of July 26, versus $7 billion last year and about $2.2 billion two years earlier [2]. The company has committed more than $40 billion in 2026 alone, according to the same Friday report on the investment surge [1].

Nvidia’s Chief Financial Officer Colette Kress framed part of the rationale around the funding needs of frontier AI labs. She said these labs have “extraordinary” demand for training and inference compute but are growing faster than their balance sheets and credit profiles can support [2]. That statement is central to understanding the strategy: Nvidia is not merely investing for financial return; it is helping ensure that the most compute-hungry firms can keep buying, leasing or accessing the infrastructure on which their models depend.

The biggest bucket highlighted in the reporting is frontier AI labs. Nvidia has invested nearly $50 billion in those labs, and OpenAI was named among the most significant recipients after a $30 billion Nvidia commitment tied to a larger funding round earlier this year [1]. If AI labs are the engines of model demand, Nvidia’s capital is acting as fuel for the engines that, in turn, require more Nvidia compute.

A second bucket is the neocloud segment: companies that buy large quantities of GPUs and rent compute capacity to enterprises and developers. CoreWeave and Nebius were cited as recipients of multibillion-dollar Nvidia commitments [1]. These companies are particularly important because they convert chip demand into a service model, giving end users access to Nvidia-powered capacity without having to build full data centers themselves.

The strategic logic: funding the flywheel

Nvidia’s AI investment strategy is best understood as a flywheel. AI labs and cloud providers need chips and networking systems. Those systems require large upfront capital commitments. Nvidia has the balance sheet, cash generation and strategic incentive to reduce that financing bottleneck. If the funded companies scale, they buy or lease more Nvidia infrastructure, which reinforces Nvidia’s revenue base and platform position.

That logic is visible in the quoted CFO response reported by MT Newswires: for companies whose models improve with more compute, more compute can mean more intelligence, more users and more revenue, making Nvidia’s participation a way to power the cycle [2]. The risk, however, is that the same cycle can look circular. When a supplier finances customers or adjacent firms, investors must ask whether demand is purely organic, partly vendor-enabled, or both.

The company’s recent MediaTek move shows another side of the strategy. Nvidia invested $3.5 billion in a zero-coupon bond issued by MediaTek as part of a broader convertible-bond offering, according to Axios [4]. The deal is tied to MediaTek’s adoption of Nvidia’s NVLink Fusion platform, which is designed to let custom AI processors connect into Nvidia’s broader rack-scale systems [4]. That matters because hyperscalers and model developers increasingly want custom silicon. Nvidia’s response is not only to compete with custom chips but to make sure those chips still plug into Nvidia’s infrastructure.

This is defensive and expansionary at once. Defensive, because it helps prevent custom accelerators from bypassing Nvidia’s networking and system architecture. Expansionary, because it potentially turns Nvidia’s platform into the default fabric for a wider range of AI processors. If the GPU is no longer the only chip in an AI factory, Nvidia wants the factory’s roads, switches and operating logic to remain Nvidia-controlled.

Hugging Face and the application layer

The reported Hugging Face transaction adds a different dimension. Reuters, through MarketScreener, reported that Nvidia would buy Hugging Face for $12.93 billion in a deal aimed at strengthening its position in open-source AI models [3]. The transaction would bring Nvidia closer to developers who use Hugging Face to find, share and deploy AI models, potentially creating a pipeline of future customers for Nvidia processors and software [3].

That is not the same as buying a cloud provider or financing a model lab. Hugging Face sits at the developer and model-distribution layer. Its importance lies in influence: where developers test models, compare tools and build workflows. By moving into that layer, Nvidia would be extending its reach beyond infrastructure into the community where many AI workloads begin.

The deal also highlights Nvidia’s need to stay relevant across open and closed AI ecosystems. Frontier labs such as OpenAI and Anthropic are often associated with closed models, while Hugging Face is identified with open-source and open-weight development [3]. A hardware platform that wants to serve the whole market benefits from influence on both sides.

SoundHound and the widening AI map

The AI equity story is not only about model labs and data centers. It also reaches application companies that turn AI into customer-facing tools. SoundHound AI announced Friday that it had completed its acquisition of LivePerson and appointed John Collins as CFO of the combined company [5]. The company said the combination expands its footprint to customers including 25 of the Fortune 100 and strengthens its intellectual-property portfolio to more than 750 patents [5].

This development is relevant to Nvidia’s broader AI-investment context because it illustrates how capital and consolidation are moving through the application layer as well. SoundHound combines voice and agentic AI with LivePerson’s digital messaging infrastructure, aiming to build an omnichannel customer-engagement platform [5]. Even if the largest dollar amounts in Nvidia’s portfolio are concentrated in infrastructure and frontier labs, the broader ecosystem includes software firms trying to commercialize AI in contact centers, enterprise workflows and consumer interactions.

SoundHound also said LivePerson’s platform would be integrated into OASYS, its self-learning Orchestrated Agent System, and that the combined company would operate across voice, web, mobile, SMS and social channels [5]. That kind of integration shows why Nvidia’s investment map cannot be understood only as a list of chip buyers. AI demand is being created by end applications, and those applications eventually feed back into compute requirements.

Why investors are paying attention

The $99 billion figure is significant enough to force a governance and valuation debate. On one hand, Nvidia is using its financial power to accelerate an industry buildout that may otherwise be constrained by customer funding limits. On the other, the company is becoming financially intertwined with firms that may also support its future revenue growth.

The positive interpretation is straightforward: Nvidia is allocating capital where it has unique visibility. It knows which companies need compute, which infrastructure providers can scale, and which technology partners can extend the platform. If those companies succeed, Nvidia benefits through investment gains, chip sales, networking demand and ecosystem lock-in.

The cautious interpretation is equally important. Supplier financing can blur the line between end-market demand and capital-assisted demand. Large equity stakes can increase exposure to valuation swings. Investments in customers or near-customers can raise questions about concentration, related-party economics and the durability of revenue if funding conditions tighten. Recent reporting noted investor concerns that Nvidia’s growing investments, sometimes involving its own customers, could inflate valuations or contribute to bubble dynamics in AI [3].

For now, the market is being asked to value Nvidia as more than a semiconductor company. It is an infrastructure platform, an AI financier, a strategic investor and, increasingly, an ecosystem architect. The $99 billion number makes that transformation visible.

The bottom line

Nvidia’s AI equity investments hitting $99 billion in one year is not just a financial statistic. It is a signal that the center of gravity in AI has shifted toward companies that can combine technology leadership with balance-sheet power. Nvidia is using capital to secure demand, shape technical standards, support customers, enter new layers of the AI stack and defend its role as the platform behind the boom.

That strategy may prove highly effective if AI adoption keeps accelerating and funded partners convert capital into sustainable revenue. It may also magnify risks if AI valuations fall, infrastructure demand slows or regulators and investors scrutinize circular financing more aggressively. Either way, Nvidia is no longer merely selling picks and shovels in the AI gold rush. It is financing miners, building roads to the mines and buying stakes in the towns that form around them.

Developments

1. Nvidia built a $99 billion equity portfolio from almost scratch in 2 yearsBusiness Insider · Sep 4, 2026, 2:22 PM UTC · 7/10
2. Nvidia's AI Equity Investments Hit $99 Billion in One Yearqz.com · Sep 4, 2026, 1:25 PM UTC · 9/10
3. Nvidia AI Equity Investments Hit $99 Billion in One Yearqz.com · Sep 4, 2026, 1:25 PM UTC · 9/10
4. Nvidia's strategic tech investments hit $99 billion as backing growsCNBC · Sep 4, 2026, 8:08 AM UTC · 9/10

Sources from the last 72 hours

1. [1]Nvidia AI equity investments hit $99 billion in one yearSep 4, 2026, 1:25 PM UTC
2. [2]Update: Market Chatter: Nvidia Increases Equity Investments to $99 BillionSep 4, 2026, 10:27 AM UTC
3. [3]Nvidia bets $13 billion on open AI models with Hugging Face dealSep 3, 2026, 12:08 PM UTC
4. [4]AI frenzy means that debt can be interest freeSep 2, 2026, 11:15 AM UTC
5. [5]SoundHound AI Completes Acquisition of LivePerson, Creating a World-Leading Omnichannel Conversational AI PowerhouseSep 4, 2026, 1:26 PM UTC

AI-generated article based on recent web research, then preserved as a dated editorial snapshot.