This race rests on a fundamental contradiction between soaring capital expenditures on graphics processing units and the sluggish pace of monetizing new products. Financial markets and technology executives face a stark operational question: can companies build out massive artificial intelligence infrastructure before their legacy cash flows collapse under the weight of these investments?
The pursuit of a trillion-dollar computing core threatens the core cash flows of Big Tech as AI product monetization lags behind record capital outlays. Investors are offloading shares of hyperscalers, driven by mounting anxiety that traditional business segments cannot generate sufficient free cash flow to cover unprecedented hardware costs. At the same time, leaders across major firms continue to lock in aggressive growth targets.
Debt and Advertising Deceleration
The economics of this transition collide with a slowdown in primary revenue drivers as user behavior shifts. Search queries handled through new conversational interfaces take longer to process and yield lower advertising returns than traditional results pages.
Look, in Q3, Search grew 12% year-over-year to $63b… it’s worth noting that the strong growth in AI Mode queries had an impact because of the lower monetization we were seeing in AI Mode.
Addressing investor concerns, Amazon CEO Andy Jassy noted that agentic purchasing through external horizontal agents has yet to meaningfully impact the company’s advertising revenue, as only a small fraction of customers currently complete transactions via autonomous paths.
Infrastructure Scenarios and Market Resilience
As ad revenue growth slows, the market positions of individual players diverge. Microsoft stands out as the only hyperscaler whose enterprise franchise is expanding average revenue per user through artificial intelligence faster than seat-based job growth decelerates. At the same time, market participants are voicing concerns over the potential displacement of traditional per-user subscription models.
Projections indicate that if current growth rates persist with an annual 15% decay, Google Cloud could surpass Azure by mid-2027 and outpace Amazon Web Services by the end of that same year. Google Cloud more than doubled its revenue by 103% to $30.7 billion, fueled by API token volumes reaching 32 billion per minute and an expanding cloud backlog totaling $561 billion. Meanwhile, Azure grew 48% to $33.2 billion, and Amazon Web Services accelerated to a 39% growth rate with $46.0 billion, marking its fastest pace since the fourth quarter of 2021. The transition toward a trillion-dollar infrastructure ultimately hinges on whether this new computing engine hits its financial targets before legacy business models exhaust their financial buffer.
Why it matters
Hyperscaler capital expenditures are breaking records, yet investors are dumping their shares amid sluggish returns from the new technology. In late September, Big Tech was already heading toward $1.2 trillion in infrastructure spending by 2027, and traditional cash flows are feeling the squeeze as conversational search queries take longer to process and yield lower ad revenue. The whole buildout is turning into a battle of endurance between cloud divisions, where Google Cloud is leaning on a $561 billion backlog to mount a direct challenge against its rivals. Current momentum points to a projected crossing where Google Cloud overtakes Azure by mid-2027.
Our stories: Big Tech $1.2T AI Spending Plan Outpaces Revenue Growth (September 27).