The Bank of England identified a potential link between energy costs and AI companies' share prices in its recent survey of financial risks, warning that the war with Iran could exacerbate pre-existing market fragilities, weigh on growth, increase inflation and tighten financial conditions.

"Prior to the conflict, increasing debt-financing needs and concerns about whether expected returns on very significant AI-related investments would materialise led to selling pressure," the Bank of England Financial Policy Committee wrote. "The conflict could increase these concerns, particularly given the energy-intensive nature of the supply chain for key components and the operation of datacentres."

The AI industry uses a large amount of energy, and the sector's rapid expansion has made it particularly exposed to rising costs. In the first three quarters of last year, 70 percent of investment growth in the United States was in AI-related goods. According to a Quinn Emanuel note, the AI sector's revenues last year were about $60 billion and its capital expenditure was $400 billion.

Robert Staiger, chief economist of the World Trade Organization, said: "The boom is very energy intensive." He added: "A prolonged period of high energy prices could crimp investment in the sector."

Hyperscalers and infrastructure providers such as CoreWeave have borrowed large amounts to build data centers. Many AI-related investments are financed by large amounts of debt. Some technology companies have issued bonds to finance AI investments, while lenders to AI companies are often private entities such as asset managers. Regulators including the Bank of England have warned about the opacity of the private credit sector financing AI investments.

Data center operators have created off-balance-sheet special purpose vehicles that own data centers and their future rental income and borrow against them. AI companies also use asset-backed securities in their financing. According to Quinn Emanuel analysts, about $120 billion in data center debt has been moved off balance sheets in the past two years. Some data center-related debts are pooled, tranched and sold to pension funds and investment managers.

According to Quinn Emanuel analysts, the interconnected AI ecosystem means distress at any single node can propagate across multiple counterparties and financing layers. Financing structures using special purpose vehicles can obscure liabilities and create a mistaken impression that risks are spread rather than concentrated. Technology analyst Ed Zitron said real-world AI infrastructure projects lag far behind companies' promises.

The energy shock from the conflict has reverberated across the globe. Many oil-importing economies, especially in the global south, are contemplating outright shortages of oil and its products, and the Philippines has declared a national energy emergency. While the U.S. is a wealthy oil exporter, gasoline prices in the country are increasing rapidly.