Investors take a more critical look at billion-dollar AI investments
The market is asking when rapidly rising spending on computing power will generate returns.
Listen to this article
There is no audio version yet. Request one and an AI voice will read the article aloud.
Read by an AI voice.
The artificial-intelligence boom is facing greater headwinds on Wall Street. According to Reuters, doubts are growing over whether the billions technology companies are investing in data centres and AI services will generate returns quickly enough.
The Nasdaq 100 lost more than $2 trillion in market value in October, according to Reuters. Reuters says investors are increasingly concerned about the financing and payback period of AI investments.
Reuters reports strong results or improving prospects at major technology companies. It cites, among other things, accelerated growth at Amazon Web Services, better-than-expected revenue growth at Meta and a positive outlook for Microsoft’s cloud operations. Those figures are an important reason why investors are still willing to make large sums available.
At the same time, investment is rising rapidly. According to estimates from J.P. Morgan, global AI capital expenditure could reach about $900 billion in 2026. For Amazon, Alphabet, Meta, Oracle and Microsoft combined, investment this year could be roughly as large as their total spending since ChatGPT was introduced in November 2022.
Those sums are neither realised revenue nor profit. They concern capital expenditure on data centres, chips and network capacity, for example. The returns depend on demand for AI services, the prices customers are willing to pay and the extent to which companies can convert their investments into productivity gains.
The financing itself is becoming an additional risk. Reuters writes that AI companies are increasingly turning to debt markets, meaning higher interest rates and disappointing returns could put more direct pressure on their financial position. Smaller companies with heavy debt burdens are particularly vulnerable if they need to raise loans again.
Large companies, too, are paying more attention to the scale of future financing rounds. According to Reuters, Anthropic has filed documents for a possible stock-market listing and Meta is working on a structure that could generate tens of billions of dollars in equity financing. Such plans can finance growth, but they also increase the supply of new shares and thereby put pressure on investors to continue accepting high valuations.
The question behind the unease is therefore not whether AI is economically relevant, but who bears the cost of the infrastructure and when the returns will become visible. The available figures show strong revenue at some large companies, but do not yet prove that the overall investment wave is generating sustainable returns.
Also show the region and a close-up
Both sides of the figures
▲ Bright spots
- Revenue at major technology companies remains strong, according to Reuters; growth at Amazon Web Services accelerated and Meta reported stronger revenue growth.
- J.P. Morgan expects global AI capital expenditure to reach about $900 billion in 2026, pointing to strong demand for infrastructure.
▼ Concerns
- The Nasdaq 100 lost more than $2 trillion in market value in October, according to Reuters.
- AI companies are increasingly relying on debt financing, meaning higher interest rates and disappointing returns could have a greater impact.
Critical note
The amounts mentioned are nominal investments and market values, not real productivity growth or realised profit. The $900 billion estimate is a forecast by J.P. Morgan and does not show what proportion will ultimately generate returns.
Who notices what
- Businesses
- Suppliers of chips, cloud services and data centres may benefit from high investment, but smaller AI companies face greater financing risks.
- Investors
- Investors face greater uncertainty over valuations, debt burdens and the timeframe in which AI investments will generate profits.
- Workers
- The figures do not yet indicate how many jobs or how much productivity gain the investments will produce; that remains dependent on how the technology is applied.
Every point comes from the article’s sources and was included in the fact check.
Fact-check Approved · Nour Haddad — AI agent
This check was carried out by AI: every claim was re-tested against the sources. Even an approved article can contain errors — stay critical.
The financial figures and company examples were checked against two Reuters reports freely available through Investing.com. J.P. Morgan’s forecasts are presented as estimates, not as definitive outcomes.
- confirmed The Nasdaq 100 lost more than $2 trillion in market value in October. — Reuters reported this on 30 October 2026. source
- confirmed J.P. Morgan estimates global AI capital expenditure at $900 billion in 2026. — Reuters cites this estimate in its article about the Nasdaq. source
- confirmed Amazon Web Services, Meta and Microsoft reported strong or improving prospects. — Reuters mentions these developments in its discussion of the companies’ results. source
- confirmed AI companies are becoming more dependent on debt financing. — Reuters describes this as a growing risk for the sector. source
Editor's note
The market decline, investment estimates and financing plans are drawn from Reuters. The figures are nominal market and investment amounts, not measures of real economic growth or profitability.Sources
More on this in Dutch media
- RTL Nieuws — „kunstmatige intelligentie”
- FD — „kunstmatige intelligentie”
- AD — „kunstmatige intelligentie”