Nvidia's role as the central bank of AI raises financial stability questions

The central bank of AI

Nvidia has positioned itself as a foundational pillar for the artificial intelligence industry, earning the nickname the 'central bank of AI.' The company has pledged hundreds of billions of dollars in financial promises to various other firms to help the industry grow. These obligations are designed to ensure that the infrastructure for AI continues to expand rapidly.

One specific example of this support is Nvidia's commitment to buy up to $36 billion in cloud services from smaller firms known as neoclouds. These neoclouds build data centers using Nvidia chips and then lease that capacity to other companies. By backing these firms, Nvidia helps them find easier financing while ensuring it has a steady stream of customers for its hardware.

However, these arrangements are increasingly viewed by some finance officials as 'circular financing.' This occurs when a large firm helps its customers buy its own products through financial assistance. While this can be productive during periods of growth, it creates a complex web of dependencies that could become problematic if the market turns sour.

Systemic risks and debt

Financial regulators, including the Federal Reserve, are closely monitoring the AI buildout to watch for potential destabilization of the global financial system. They are concerned that many factors helping the AI boom could reverse simultaneously. For instance, a drop in the value of Nvidia's GPU chips could prompt investors to worry about whether companies that borrowed money to buy them can repay their debts.

The scale of tech involvement in debt markets has grown significantly in a short period. Large technology firms have increased their share of the $15 trillion investment-grade debt market from a mere 1% to 10% in just two years. Additionally, hedge funds and retail investors are taking on debt to buy stocks or investing in funds that use embedded leverage.

If tech companies face credit problems at the same time their equity values fall, the impact would not stay contained to tech portfolios. Jason Wu of the International Monetary Fund warned that because the economy and financial markets are standing on the same leg, a drawdown could have reinforcing effects between the two. This suggests that a crisis in the AI sector could ripple across the broader economy.

Nvidia's financial cushion

Despite these concerns, many experts believe Nvidia is not 'too big to fail' in the same way a traditional bank might be. Sheila Bair, who previously served as a government regulator leading the Federal Deposit Insurance Corp., noted that Nvidia is not vulnerable to a sudden loss of confidence. Because it produces a physical product that customers want to buy, it would likely remain viable even if it faced significant corporate trouble.

The company also possesses massive equity, which provides a significant cushion against financial shocks. Stijn Van Nieuwerburgh, a Columbia Business School finance professor, estimates that Nvidia's total obligations are around $400 billion. While this is a large sum, he notes it is a small fraction of the $3 trillion market capitalization the company holds even after recent stock declines.

Van Nieuwerburgh suggests that the company would have to reach the brink of extinction before all of these guarantees would come into jeopardy. He noted that while the system is not perfectly safe, there is currently a lot of equity in the system. This suggests that while the risks are real, the immediate threat of a total collapse is mitigated by the company's current financial standing.

Market volatility and competition

Nvidia's stock has experienced notable volatility recently, falling 30% from its October 6 peak through a recent Tuesday close. This move was part of a broader reassessment of the AI trade following the November 3 midterms. However, the stock recovered quickly, rising more than 4% as of midday on Wednesday, showing that investor sentiment remains somewhat resilient.

The company is also facing internal and external pressures, including scrutiny over its CEO, Jensen Huang. Reports have surfaced regarding his relationship with a woman suspected of spying for China, leading to potential calls for his removal. While the board has expressed confidence in Huang, these allegations add a layer of uncertainty to the company's leadership and stability.

In the long term, competition may act as a buffer against the economic fallout of any problems at Nvidia. Several major players, including AMD, Broadcom, Intel, Google, Microsoft, and OpenAI, are all working to build their own AI chips. Van Nieuwerburgh noted that these efforts could eventually replace Nvidia's products, potentially capping the damage if the company faces a significant downturn.