A former UBS chief investment strategist warns that AI CapEx (capital expenditure) is accelerating at a record pace, lifting valuations and concentrating power in a handful of technology giants.
In a post published in the Financial Times, Ian Harnett says the same forces that drove the boom may now be signaling its top — echoing the last great tech cycles that ended in collapse.
He points to runaway spending among hyperscalers and a narrowing of stock leadership across major US indexes as symptoms of excess.
“The AI ‘bubble’ looks to be approaching its endgame. The dramatic rise in AI capital expenditure by so-called hyperscalers of the technology and the stock concentration in US equities are classic peak bubble signals.”
The exuberance, he says, is being amplified by circular deal-making inside the sector itself. Companies are inflating valuations through mutual purchases and vendor financing, creating the illusion of organic growth.
Harnett believes the final stage of the cycle began when tech firms moved from software hype to real-world hardware expansion. The multibillion-dollar race to build data centers and compute infrastructure, he argues, mirrors the 1990s TMT (telecommunications, media, and technology) boom that fueled the internet’s rise — and its eventual collapse.
“Until recently, the missing ingredient was the rapid build-out of physical capital. This is now firmly in place, echoing the capex boom seen in the late-1990s bubble in telecommunications, media, and technology stocks.”
He warns that the trigger for reversal may not come from within the tech sector, but from the buyers themselves. If enterprise clients or consumers experience a cash-flow disruption, the entire cycle could unwind faster than companies can cut back on spending.
“If the end buyers of AI suffer an exogenous cash flow shock, the merry-go-round will slow rapidly as sales collapse faster than capex can be reined in, resulting in faltering earnings and accelerated cash burn.”
For investors, Harnett’s warning carries a reminder from history. Even the eventual winners of the dot-com era suffered catastrophic drawdowns before recovering years later.
“Microsoft (65%), Apple (80%), Oracle (88%), and Amazon (94%). These companies took 16, five, 14, and seven years respectively to regain their 2000 peaks.”
Still, he doesn’t dismiss the long-term promise of AI. Like past technological revolutions, he believes today’s excess will eventually finance the infrastructure that makes future innovation possible — though investors may endure a painful reset along the way.
“The good news is that the aggressive AI CapEx build-out almost guarantees AI’s future ubiquity. The bad news is that this will probably only come after a period of Schumpeterian creative destruction.”
Disclaimer: Opinions expressed at CapitalAI Daily are not investment advice. Investors should do their own due diligence before making any decisions involving securities, cryptocurrencies, or digital assets. Your transfers and trades are at your own risk, and any losses you may incur are your responsibility. CapitalAI Daily does not recommend the buying or selling of any assets, nor is CapitalAI Daily an investment advisor. See our Editorial Standards and Terms of Use.