Summing up the bear case for AI
The artificial intelligence (AI) industry is currently grappling with what some experts call a ‘trillion-dollar math problem’. The numbers might not stack up because customers might simply lack the funds to spend on AI tools to allow hyperscalers to achieve a decent return on their AI infrastructure investment.
With hyperscalers projected to spend US$3 trillion on AI infrastructure by 2029, the market faces a substantial revenue gap. To justify current valuations and maintain reasonable margins, AI services would need to generate revenue equivalent to 10 per cent of the entire U.S. Gross Domestic Product (GDP) of US$30 trillion. This represents a massive commercial risk; if expectations of an adequate return on investment in two or three years evaporate, this historic capital expenditure risks producing a multi-trillion-dollar overcapacity.
Fears that structural spending could meet cyclical realities would generate a period of creative destruction.
Historically, when periods of creative destruction ended past general-purpose technology (GPT) booms, investors saw up to 90 per cent of their gains wiped out. Past general-purpose technologies, including the automobile, electricity, and the internet, have indeed changed the world, but they also simultaneously destroyed the capital of their earliest backers.
During the Dotcom bubble, even eventual winners such as Amazon and Microsoft saw their share prices collapse by more than 90 per cent when the hype met the cyclical reality. The core issue is that while the AI theme is structural, customer demand remains cyclical. When a structural dream meets cyclical commerciality, the cyclical reality has won every time. The eventual winners are those who buy the ‘creatively destroyed’ distressed assets from the original, distraught investors at cents on the dollar.
In fact, it is through distressed-asset buyers that the technology can be widely and globally distributed at affordable prices, ensuring the technology changes the world.
Elsewhere, and putting aside the immediate capital expenditure issues, concerns are mounting over what some observers call depreciation games, prompting warnings that tech giants could be hiding the true operating costs of supplying AI tools through accounting manoeuvres. Specifically, hyperscalers are said to be extending the depreciation schedules of their chips from three years to six, even as the hardware upgrade cycle accelerates with annual releases like Nvidia’s Blackwell. Famed short-seller Michael Burry has flagged this as an accounting hoax designed to mask the rapid obsolescence of hardware and artificially inflate profits.
Meanwhile, this build-out is increasingly being financed with debt. Nearly 10 per cent of recent corporate bonds have been issued specifically for AI projects. Most recently, Firmus, an Australian company that rents and builds large data centres earmarked for use by OpenAI, Anthropic, Meta, Amazon, and Microsoft, secured a US$10 billion financing package led by the private equity firm Blackstone. Not to be outdone, Alphabet Inc. (Google) initiated a massive debt offering to fund its artificial intelligence infrastructure, drawing over US$100 billion in demand for a bond sale expected to be around US$15 to US$20 billion.
This massive oversubscription highlights the intense investor demand for debt tied to the AI boom, with Alphabet even exploring rare 100-year bonds to finance data centre construction.
Meanwhile, we are also witnessing the disruptors being disrupted, as last week’s software slaughter accelerated a year-long sell-off, indicating AI might be eating the traditional Software-as-a-Service (SaaS) industry alive.
New autonomous agents are beginning to replace the SaaS tools that have defined the last two decades, raising genuine durability concerns for established market leaders in data and services such as Thomson Reuters and LexisNexis.
That the bell might be ringing an end to the AI boom might have nothing to do with numbers or growth at all, but sentiment. The recent crowning of the ‘Architects of AI’ on the cover of TIME Magazine in December has historically served as a harbinger of doom for stock trajectories. Much like the covers featuring Jeff Bezos in 1999 or Elon Musk in 2021, these moments of peak public interest often signal that an investment theme is saturated, and the trajectory is about to reverse.
Finally, there is an emerging efficiency threat in which cheaper, faster models from global competitors could erode the expensive U.S. AI moat. As international models prove that AI can be delivered with significantly less computing power and energy, the trillions spent on massive data centres could result in a period of Schumpeterian (creative destruction) waste. If the technology becomes a cheap commodity, any assumed competitive advantage from having the largest infrastructure will erode.
Today’s view is the ultimate winners of the AI revolution will be those currently spending the most. While a tantalising prospect, it strikes me as overly simplistic. I think it will be those that survive the collision between the structural theme and the cyclical economic reality.
Roger Montgomery is the Founder and Chairman of Montgomery Investment Management. Roger has over three decades of experience in funds management and related activities, including equities analysis, equity and derivatives strategy, trading and stockbroking. Prior to establishing Montgomery, Roger held positions at Ord Minnett Jardine Fleming, BT (Australia) Limited and Merrill Lynch.
He is also author of best-selling investment guide-book for the stock market, Value.able – how to value the best stocks and buy them for less than they are worth.
Roger appears regularly on television and radio, and in the press, including ABC radio and TV, The Australian and Ausbiz. View upcoming media appearances.
This post was contributed by a representative of Montgomery Investment Management Pty Limited (AFSL No. 354564). The principal purpose of this post is to provide factual information and not provide financial product advice. Additionally, the information provided is not intended to provide any recommendation or opinion about any financial product. Any commentary and statements of opinion however may contain general advice only that is prepared without taking into account your personal objectives, financial circumstances or needs. Because of this, before acting on any of the information provided, you should always consider its appropriateness in light of your personal objectives, financial circumstances and needs and should consider seeking independent advice from a financial advisor if necessary before making any decisions. This post specifically excludes personal advice.
