NVIDIA commits $90 billion

NVIDIA has committed roughly $90 billion to AI-related deals over the past 16 months, according to Semafor, extending the chipmaker’s role beyond selling processors into financing the companies building around them. The report said the money is being directed to AI developers, cloud providers and infrastructure suppliers. That makes the company an increasingly important source of capital inside the same ecosystem that buys and deploys its technology. (semafor.com) The scale matters because NVIDIA is coming off a fiscal year in which it reported $215.9 billion in revenue, including $62.3 billion in fourth-quarter data center revenue, giving it unusual capacity to fund partners as well as ship hardware. (newsnow.co.uk) Investors were already focused on the company’s earnings due May 20 as a test of whether AI spending remains durable. ### Where is the $90 billion going? Semafor said NVIDIA’s commitments are spread across developers, cloud providers and infrastructure suppliers rather than concentrated in one type of transaction. (semafor.com) That means the company is not only financing end-user AI labs, but also the suppliers and service layers needed to expand compute capacity. Reports describing the push said the total covers more than 145 companies over about 16 months. That suggests the figure includes a mix of investments, supply agreements, purchase commitments and strategic partnerships, not a single fund or one-time program. (investor.nvidia.com) ### Why does this look different from ordinary corporate investing? NVIDIA has been using balance-sheet strength created by the AI boom to shape the market around its chips. In March, for example, the company announced a multiyear strategic agreement with Coherent that included a multibillion-dollar purchase commitment and future access and capacity rights for optical networking products used in AI infrastructure. (semafor.com) TechCrunch reported earlier in May that NVIDIA had already committed more than $40 billion to equity investments in AI companies in 2026 alone. (edgen.tech) Semafor’s larger $90 billion figure indicates the company’s ecosystem spending goes well beyond startup stakes and includes broader commercial arrangements. ### Why are cloud providers and developers central to this strategy? NVIDIA’s largest customers are also the companies building the next layer of AI services, cloud capacity and model infrastructure. By backing those participants directly, the company can help finance demand for the systems it sells while also reinforcing its software and networking standards across the stack. (nvidianews.nvidia.com) That description is an inference based on the categories of companies named in the reporting and NVIDIA’s central role in AI infrastructure spending. TechCrunch reported in February that NVIDIA Chief Executive Jensen Huang had estimated between $3 trillion and $4 trillion would be spent on AI infrastructure by the end of the decade. (techcrunch.com) A company operating at NVIDIA’s scale can use financing commitments to secure a larger place in that buildout. ### What questions does this raise? The concentration question is already part of the story. Semafor said the deal push would tie more of the AI economy to NVIDIA’s technology, while another report said the company’s expanding web of investments and partnerships has drawn inquiries from regulators. (semafor.com) Competitors are also trying to widen the field. Recent reporting has pointed to custom-silicon efforts by hyperscalers, new AI cloud ventures and supplier-side partnerships across optics and data centers, all of which could become counterweights if customers seek alternatives. (techcrunch.com) ### What comes next? NVIDIA is scheduled to report earnings on May 20, and investors are expected to press management on AI demand, spending visibility and the durability of the current buildout. Any discussion of partner financing, purchase commitments or ecosystem investments is likely to receive close attention alongside revenue and guidance. (semafor.com) The next hard datapoints will come from NVIDIA’s earnings materials and executive commentary, which should show whether the company frames these commitments as strategic investments, supply agreements or another category of capital deployment. (nvidianews.nvidia.com) (investor.nvidia.com) (newsnow.co.uk)