The artificial intelligence (AI) boom may be entering its late stages, according to Capital Economics, which expects the bubble could start to burst in 2027.
"There are plenty of signs that we are now in the late stages of a bubble in AI," John Higgins, chief economic adviser for financial markets at Capital Economics, said in a report Monday. The firm expects the S&P 500 to end 2027 at 6,500, compared with 8,250 at the end of 2026, and says its eventual peak-to-trough decline could be at least 30%.
Capital Economics points to several signs behind its warning. U.S. technology capital expenditure has risen to a larger share of GDP than at the peak of the dot-com boom, while earnings expectations remain optimistic and valuations are stretched. Equity issuance has also surged, while the market has become increasingly concentrated in a small number of companies.
The firm also expects the eventual decline to be less severe than the roughly 50% fall during the dot-com bust because today's AI boom has been driven more by earnings growth than by investors simply paying higher prices for earnings.
Capital Economics, however, questions whether analysts' expectations for earnings growth can remain this strong.
Warnings Mount
That concern comes as investors have already raised questions about AI spending and valuations. J.P. Morgan Asset Management's Bill Eigen previously warned that investors should watch the rate at which AI growth accelerates and pointed to signs of slowing capital expenditure and private-market valuations.
A Bank of America survey also found that 48% of investors viewed AI hyperscaler capital expenditure as the most likely source of a systemic credit event, while 45% identified an AI bubble as the biggest tail risk.
Capital Economics senior markets economist James Reilly told CBS News that the earnings expectations for leading AI companies look stretched compared with U.S. economic growth, calling it "the dot-com bubble all over again." He said AI will be transformative and profitable, but profits may not reach analysts' expectations.