Jefferies warns that the AI capital expenditure cycle may be nearing an inflection point, with its contribution to U.S. GDP growth having reached 67%.

BlockBeats news, November 7 - Jefferies strategist Christopher Wood's latest report states that in the past four quarters ending in the third quarter of 2026, the contribution of US AI-related real capital expenditure to year-over-year GDP growth reached 1.39 percentage points, accounting for 67% of the overall 2.1% US economic growth rate. Meanwhile, investment in information processing equipment and software as a share of nominal GDP rose to 5.35%, surpassing the 4.46% during the 2000 internet bubble period and hitting a record high.

Jefferies noted that widening credit spreads, continued declines in semiconductor stocks, and deteriorating free cash flow at hyperscale cloud service providers all indicate that the AI capital expenditure cycle may be approaching an inflection point. Microsoft, Amazon, Alphabet, and Meta are expected to spend a combined total of approximately $1.04 trillion in capital expenditure this year, but Amazon and Alphabet's free cash flow has already dropped to -$35 billion and -$9 billion, respectively.

The report also warns that the AI ecosystem has formed a contract backlog and circular financing system worth approximately $2.5 trillion, with data center financing increasingly relying on private credit and off-balance-sheet financing structures. Once AI investment expectations are revised downward, it could trigger a chain reaction of tightening financing and contracting capital expenditure.

Jefferies believes that a sharp contraction in AI investment would deal a significant blow to US economic growth and could force the Federal Reserve to shift toward easing. However, since the US fiscal deficit has already reached 6.2% of GDP, US Treasuries may not necessarily play their traditional role as a safe-haven asset at that point. The report recommends that investors appropriately diversify their asset allocation beyond AI and pay attention to areas such as gold and biotechnology.