Nvidia is known as the architect of the AI ecosystem. It is now also recognized as a major financier in the field.
Nvidia has committed nearly $90 billion in financial operations and partnerships over the past sixteen months. These commitments span more than 145 companies, from AI model developers to cloud providers and infrastructure players.
As one Dawn Liphardt Valley banker succinctly put it to the Financial Times:
“They fund everyone.”
The peculiarity of Nvidia’s strategy lies in its hybrid form. While the group runs a venture capital arm named NVentures, it is the business development division that has driven the bulk of the deals. Its investments often accompany broader commercial partnerships.
The mechanism is systematic: Nvidia takes an equity stake in a company in exchange for a technical integration into its own ecosystem.
That is how the startup SiFive came to Nvidia’s capital after agreeing to render its chip designs compatible with NVLink, Nvidia’s proprietary interconnection technology.
A similar arrangement was reached with Marvell, the chip designer behind Amazon’s Trainium AI accelerators. Nvidia invested $2 billion in March, paired with a partnership to make Marvell’s future custom chips compatible with NVLink.
Investments Coupled with Commercial Partnerships
On the ground, negotiations often unfold in two parallel sessions. Nvidia’s technical teams engage with the companies, while the business development squads show up “with checks.”
The strategy extends to new cloud infrastructure providers, which Nvidia helped bring into being to counterbalance the growing power of hyperscalers—who are not only its largest customers but also potential rivals in the chip market.
The deal announced this month with the neocloud group Iren illustrates how far this interweaving of roles can go. Nvidia has committed to spending $3.4 billion over five years to lease GPU capacity, while investing up to $2.1 billion in the company. Nvidia thus finds itself simultaneously a customer, a supplier, and a shareholder.
Beyond equity, Nvidia is also pursuing a strategic push to anchor its software layer. The companies it invests in are regularly encouraged to adopt its Nemotron models.
Jensen Huang therefore hopes to replicate the success of CUDA, the ecosystem that has become the de facto standard for enterprise AI. In venture circles, the message is now well understood: “If you build on Nvidia’s ecosystem, you can raise funds with Jensen.”
Beyond its equity stakes, Nvidia has also committed $95 billion to securing its supply of components and its manufacturing capacity.
Among the latest agreements are two $2 billion investments in photonics groups Coherent and Lumentum, as well as $3.2 billion in equity warrants for Corning, the fiber maker used in high-performance data centers.
Risks Mounting
This capital frenzy does not come without trade-offs. The spending associated with transactions absorbed roughly 40% of Nvidia’s operating cash flow in its most recent fiscal year—a very high ratio compared with Alphabet’s 6%, even though Alphabet is widely regarded as the tech industry’s largest startup investor.
Regulatory authorities in the United States, the European Union, and the United Kingdom have issued broad information requests to Nvidia regarding its investments, partnerships, and deals with companies developing foundational models.
Nvidia, for its part, contends that its strategy “accelerates and diversifies innovation” and that its investments are “not conditioned on exclusivity.”
Yet for many observers, the reality is more nuanced. By funding suppliers, customers, and potential competitors at once, Jensen Huang is weaving a web in which Nvidia will find itself at the center, whatever the outcome.
