Nvidia is preparing a roughly $20 billion bond sale in the United States, according to a person familiar with the matter, tapping the investment-grade debt market to help finance the heavy capital needs associated with producing next-generation artificial intelligence chips. The planned transaction would be the company’s first such issuance in five years, with Nvidia previously raising $5 billion in June 2021, the source said, speaking on condition of anonymity because the plan remained private.
A term sheet reviewed by the source shows the offering is structured as seven separate tranches of notes, with maturities stretching as far as 2056. The document also indicates that the proceeds are intended for general corporate purposes, explicitly including repayment and refinancing of outstanding notes.
Goldman Sachs, J.P. Morgan and Morgan Stanley are listed as the bookrunners on the deal. Nvidia had not immediately provided comment when inquiries were made.
Financial data in the term sheet noted that the company held $13.24 billion in cash and cash equivalents for the quarter ended April 2026. The issuance would augment that liquidity position while addressing ongoing funding requirements tied to the company’s chip development and production roadmap.
Market reaction to the news was positive in early trading, with Nvidia shares rising about 2.5% as the details circulated.
The fundraising comes as Big Tech continues to signal sustained, and rising, investment in artificial intelligence. Combined outlays by major technology firms are set to surpass $700 billion this year, up from around $400 billion in 2025, according to figures cited in the source material. Within that context, several large technology companies have also accessed the bond market recently: Meta filed in October for a potential bond offering of up to $30 billion, and Alphabet has disclosed plans to issue Japanese yen-denominated debt for the first time.
While Nvidia itself has not been primarily focused on constructing large-scale data centers, its semiconductors are a critical component inside the servers that power AI workloads. Demand for those chips has been described as red-hot by market participants, driven by companies that need high-performance processors to train and operate increasingly sophisticated AI models.
To keep pace with the rapid evolution of AI computing, Nvidia has been investing heavily in chip design and manufacturing. The company has been releasing new families of processors on an approximately annual cadence, with each generation offering higher AI performance than the last, according to the information provided.
Contextual note - The reported bond issuance is presented as a financing step to support Nvidia’s capital-intensive push into advanced AI chip production and to manage its outstanding debt maturities. The structure, bank syndicate and stated uses of proceeds are drawn from the term sheet and the account of the unnamed source.