14 April 2026
Over the past decade, spending on artificial intelligence (AI) has represented one of the largest waves of private investment in modern history.
Just this year, the four major cloud hyperscalers – Amazon, Google, Meta and Microsoft – are expected to deliver over $500bn on capital expenditures, a 60% increase on 20251. Estimates for total data centre infrastructure private capex globally in 2026 run well into the trillions of dollars.2
Source: LSEG Workspace, IBES estimates
Such an outlay has understandably led to concerns around whether it will actually yield an appropriate return on investment (ROI) or come to be known as one of the biggest overspends in history. This can ultimately be boiled down to two simple questions:
Is the total addressable market (TAM) for AI sufficiently large?
How fast will that TAM arrive?
AI is on course to be one of the most disruptive technologies of our generation, underpinning a significant and durable investment cycle that investors cannot ignore.
But we should also pay close attention to risks that could delay the pace at which the technology is deployed and adopted. Setbacks could pose a serious threat to achieving the necessary return on investment on which current spending and valuations – as well as the broader economy – are becoming increasingly reliant.
To read more, download the full article.
This insight is part of our broader analysis on how today’s global shifts are impacting investment opportunities – a dynamic we call The Great Global Restructuring.
John Lamb is an investment specialist at Capital Group. He has 10 years of industry experience and has been with Capital Group for seven years. He holds a bachelor's degree in economics and management from the University of Oxford. He also holds the Chartered Financial Analyst® designation. John is based in London.
