Big Tech's AI spending spree puts over $2 trillion in Nasdaq market value at risk

The Nasdaq-100 is an index tracking 100 large nonfinancial companies listed on Nasdaq. It is heavily weighted toward technology companies. Its market value reflects what investors collectively believe those companies are worth. Saying more than $2 trillion is “at risk” means that value could decline if investors lose confidence in the AI spending boom.

The index has already fallen more than 5.8% in October. Alphabet suffered its largest one-day percentage fall after disappointing search advertising results and higher cash burn. Meta’s plan to raise $97.5 billion also increased fears about dilution and funding pressure. These events can lower share prices across the index.

The possible loss would exceed SpaceX’s entire market value. The figure signals the scale of the concern, not a guaranteed loss. Investors are watching whether companies can turn huge AI investments into profits quickly enough, especially as borrowing costs rise.

Big Tech’s AI spending has moved far beyond earlier infrastructure budgets. Companies traditionally spent heavily on cloud computing, advertising and e-commerce systems. The AI boom has pushed those outlays to much higher levels because companies are building new capacity while demand for data and AI services grows.

J.P. Morgan expects Amazon, Alphabet, Meta, Oracle and Microsoft to spend about as much on AI infrastructure this year as they spent altogether since Axiom launched in November 2022. Morgan Stanley expects total AI infrastructure spending to cross $1 trillion next year. J.P. Morgan forecasts $900 billion globally in 2026 alone.

The broader market could be much larger still. J.P. Morgan expects global AI infrastructure spending to reach $5.5 trillion by 2030. These figures show why investors are focused on returns: the capital commitment is expanding faster than certainty about when profits will arrive.

Meta, Alphabet, Amazon and other technology companies are increasing forecasts because the AI boom requires much more infrastructure. They are expanding the systems supporting cloud computing, advertising and other digital services. John Bilton described demand for data as “insatiable and growing,” helping explain why companies continue building capacity.

Recent earnings showed strong revenue in several businesses. Amazon Web Services grew faster, Meta reported better-than-expected revenue and Microsoft forecast strong cloud growth. Those results gave investors evidence that demand remains healthy, even as spending rises. Companies therefore have business momentum supporting further investment.

However, forecasts are increasing while the timing of payback remains unclear. Meta, Alphabet and Amazon raised spending expectations, and investors are questioning the cost and recovery period. Higher borrowing costs make the expansion more difficult to finance. The key issue is whether demand and revenues grow quickly enough to justify the new spending.

Stock prices reflect investors’ expectations about future profits. If AI investments take longer than expected to pay back, investors may decide that companies are spending too much for too little near-term return. They can then lower the prices they are willing to pay for those companies’ shares.

The article gives several examples of this pressure. The Nasdaq-100 fell more than 5.8% in October. Alphabet recorded its biggest one-day percentage fall after results disappointed investors in search advertising and showed sharply higher cash burn. Meta’s plan to raise $97.5 billion raised fears about shareholder dilution and funding needs.

A slower payoff could spread beyond individual stocks. Higher borrowing costs would increase expenses, while debt-funded investment could pressure credit markets. Microsoft and SpaceX have been exceptions, with results suggesting returns are beginning to appear. Still, investors are watching Apple and Nvidia for evidence that AI spending is producing durable growth.

AI infrastructure can generate revenue by supporting services that customers pay for. For Microsoft and Amazon, the clearest route in the article is cloud computing. Companies and other users can consume computing capacity through cloud services, creating revenue as demand grows. Meta’s results show a different support base: stronger company revenue can help fund its AI expansion.

Amazon Web Services continued to grow faster in the latest results. Microsoft forecast strong growth in its cloud business. Meta reported better-than-expected revenue growth and gave an upbeat outlook. These results helped show investors that AI-related infrastructure may support existing businesses rather than produce returns only through a separate AI product.

The mechanism is not yet guaranteed. The companies are spending heavily before the full financial return is clear. Strong demand and revenue offer support, but higher costs, borrowing and uncertain recovery periods create risk. Investors are therefore comparing cloud and business growth with the scale of new AI spending.

Debt-funded AI expansion requires companies to repay borrowed money and cover its financing cost. When borrowing costs rise, the same investment becomes more expensive. A project must therefore generate more revenue or profit just to reach the same financial result. Delayed returns make that gap wider.

Microsoft is reportedly planning a roughly $25 billion US investment-grade bond sale, its first in more than nine years. Meta, Amazon and Alphabet have also issued hundreds of billions of dollars in bonds over the past four years. These actions show why investors are examining how much AI expansion depends on debt.

The risk grows if AI infrastructure takes longer to pay off. Higher interest costs could reduce cash available for operations or future investment. The pressure can also reach credit markets, where investors assess whether companies once viewed as insulated could face broader economic strain. Morgan Stanley expects AI infrastructure spending to exceed $1 trillion next year.

AI infrastructure means the physical and technical systems used to develop and run AI services. It includes computing equipment, specialized processors, data centers, networks, storage and power. The article mainly uses the term for the systems behind expanding cloud and AI capacity. This explanation of the components uses established computing knowledge beyond the article’s detail.

AI workloads require large amounts of computing and data. Companies must buy or lease equipment, build facilities, connect them through networks and keep them operating. Those systems are costly even before a customer or advertiser produces revenue. That creates a long period between spending and possible payback.

The scale is now central to the market debate. Morgan Stanley expects AI infrastructure spending to exceed $1 trillion next year. J.P. Morgan forecasts $5.5 trillion globally by 2030. Companies are increasing capacity because demand is strong, but investors worry that profits may arrive more slowly than the construction bills.