TLDRs:
- Nvidia commits up to $100B to OpenAI, primarily for leased GPUs powering new AI data centers.
- OpenAI leases Nvidia chips, spreading costs but raising questions about financial sustainability.
- Analysts warn AI investment cycles may inflate valuations without generating tangible returns.
- OpenAI aims to scale AI infrastructure while balancing cost, demand, and investor scrutiny.
Nvidia’s landmark investment in OpenAI, reaching up to $100 billion, marks one of the most significant financial commitments in the AI sector.
While the headline number is staggering, most of the funds are slated for the use of Nvidia’s own graphics processing units (GPUs), leased over time to OpenAI. The first tranche of $10 billion will soon be available to support the launch of new AI supercomputing facilities, beginning with a gigawatt-scale data center in Abilene, Texas, slated to come online in late 2026.
According to Nvidia CEO Jensen Huang, building a data center of this scale could cost around $50 billion, with roughly $35 billion devoted to GPU hardware alone. By leasing rather than outright purchasing these chips, OpenAI spreads the financial burden over several years, giving the company greater flexibility while placing a degree of risk on Nvidia.
Leasing Chips to Manage Costs
OpenAI’s approach of leasing GPUs rather than buying them outright is a strategic move aimed at reducing upfront capital expenditures.
“Equity is the most expensive way to fund data centers,” said a source familiar with the arrangement.
Leasing allows OpenAI to deploy state-of-the-art GPUs immediately, while paying over time as the infrastructure becomes operational.
This model also has broader implications. By committing to long-term lease agreements, Nvidia ensures a steady revenue stream from its chips, while OpenAI retains access to the processing power necessary to train its large language models and deliver AI services like ChatGPT. Oracle is also partnering with OpenAI on facility leasing, highlighting a collaborative approach to addressing the global shortage of AI compute capacity.
Investor Concerns on Circular Funding
Despite the enthusiasm surrounding the deal, some analysts have expressed caution. Jamie Zakalik, a Neuberger Berman analyst, noted that the arrangement exemplifies the “circular nature” of AI financing, where capital injected into a company immediately flows back to its suppliers.
While this structure inflates revenues and valuations for both Nvidia and OpenAI, it may not generate new, tangible economic value.
With Nvidia approaching a $4.3 trillion market capitalization, much of its growth has been fueled by GPU sales to tech giants and startups alike. OpenAI’s own valuation has surged to nearly $500 billion, driven by investments from Microsoft and others. Critics argue that relying on massive, recycled investments raises questions about the long-term sustainability of the AI boom.
Balancing Growth and Sustainability
OpenAI CEO Sam Altman has emphasized that the company remains focused on meeting real demand.
“We need to keep selling services to consumers and businesses, and building these great new products that people pay us a lot of money for,” Altman said.
Revenue from AI services is intended to offset the costs of expensive GPUs and sprawling data centers.
The Abilene data center offers a glimpse into OpenAI’s ambitious infrastructure plans, but scaling AI to global demand will require careful management of capital, resources, and partnerships. Nvidia’s GPU leasing model and OpenAI’s use of equity funding demonstrate the delicate balance between growth, operational costs, and investor expectations in the fast-moving AI landscape.
