Nvidia's AI Financing Faces First Major Stress Test

Nvidia's extensive financing of the global AI infrastructure is encountering its initial major challenge as credit markets tighten for its strategic partners. Beyond manufacturing advanced AI chips, Nvidia, under CEO Jensen Huang, has extended its influence by providing significant financial backing and facilitating loans for AI model developers and data center companies.

This strategy leverages Nvidia's strong credit rating to secure lower interest rates for mega-projects, such as a proposed 10-gigawatt OpenAI data center in Ohio. As of July, Nvidia reported $36 billion in off-balance-sheet AI cloud agreements, with guarantees extending up to $105 billion for the OpenAI facility alone.

Market conditions are now testing Nvidia's financial strategy. Shares of neoclouds like CoreWeave, Nebius, and IREN have dropped by at least 30% this month. CoreWeave, a significant entity in Nvidia's portfolio, has seen its credit default swap spreads surge past 11%.

Nvidia's financial commitments are substantial, with over $500 billion in off-balance-sheet agreements outlined through fiscal 2029. This includes a partnership on a $500 billion financing initiative to treat compute infrastructure as an asset class, offering financial firms guarantees on GPU financings.

Despite market jitters and concerns about the stability of future contracts for neoclouds, Jensen Huang has expressed confidence in continued high demand for AI services. He noted a scarcity of working capital to fuel this growth, positioning Nvidia's AI factory platform as an "investable asset."

Nvidia is actively seeking insurers to mitigate the risks associated with potential GPU value depreciation, though insurers are not yet a complete solution. Off-balance-sheet liabilities, particularly lease obligations, have grown substantially across the AI sector, prompting close monitoring from ratings agencies like Moody's.