Nvidia dismisses "circular financing", says every $1 it invests brings back $100

The cost of building AI infrastructure significantly increases for companies that rely on Nvidia's chip supply and financing.

Nvidia CEO Jensen Huang defended the company's investments in the AI ecosystem, stating that for every dollar Nvidia invests, it receives a hundred dollars back. He dismissed concerns that these investments amount to circular financing, emphasizing that Nvidia's capital input unlocks significantly larger returns and chip demand. Nvidia has reportedly invested around $50 billion in AI labs and provided guarantees of up to $105 billion for OpenAI's data center project, where it serves as the exclusive chip supplier. A platform is also being established to mobilize approximately $500 billion in third-party capital. Despite these assertions and a reiterated Buy rating from Goldman Sachs with a $300 target, Nvidia's stock has experienced a recent decline, influenced by broader market trends like rising Treasury yields and oil prices.

Nvidia's strategy of funding AI labs, guaranteeing data center financing, and acting as an exclusive chip supplier creates a powerful commercial flywheel. This interconnectedness accelerates the AI buildout and drives demand for its GPUs, with the company projecting a 70% revenue growth. However, this model's risk lies in a potential slowdown of AI infrastructure spending or tighter financing terms, which could disrupt customer expansion and thus Nvidia's revenue trajectory. This situation also poses risks to highly leveraged cloud operators who depend on strong utilization and pricing to service debt, making them vulnerable if AI spending cools.

The significant scale and interconnectedness of Nvidia's AI investments differentiate it from traditional supplier financing models seen in aerospace or telecoms. By investing in AI companies, supporting their data center financing, and supplying the core GPUs, Nvidia is deeply entwined with its customers' growth. This creates a potential vulnerability where the demand for AI infrastructure, and by extension Nvidia's chips, becomes increasingly intertwined with the availability and terms of capital. Concerns about hyperscalers over-ordering chips due to supply scarcity also complicate the interpretation of demand signals, suggesting that current order volumes may not perfectly reflect future needs.