The market is wary of massive AI capital spending by tech giants. Unlike traditional infrastructure with long lifespans, AI chips age quickly. This creates a risk that companies will overspend on hardware that becomes obsolete before generating sufficient returns, leading to underperformance.

When stocks of major cloud providers (hyperscalers), who are the primary buyers of AI chips, lag behind the stocks of their semiconductor suppliers, it signals potential trouble. This divergence suggests the market is questioning the pace of capital spending.

The AI investment frenzy is sustained by massive capital expenditure (CapEx) on data centers. According to a Goldman analyst, the first major tech company to announce a pullback on this spending will be the key signal that the debt-fueled boom is unsustainable, potentially triggering a broader market correction.

Amazon's stock fell despite strong AWS growth because its $200B capital expenditure plan signaled the enormous cost of competing in AI. The market views this massive spending less as a guaranteed growth driver and more as a defensive necessity to keep pace, compressing margins and worrying investors.

Massive capital expenditure projections from major technology companies do not necessarily indicate real economic output. Disclosures from hyperscalers like Amazon and Microsoft indicate that significant portions of their capital outlays are merely being absorbed by soaring memory chip prices and semiconductor supply constraints rather than higher physical build volumes or real economic productivity.

Amazon is taking about $5B of Nvidia GB300 racks in August and $3B monthly after, but Business Insider reported most sit in warehouses because data centers lack power. Ed Zitron estimates $200-$300B of AI chips industry-wide are sitting dark. This shows AI buildout is constrained by energy, not chip supply.

There's a contradictory market sentiment regarding AI investment. Hyperscalers like Amazon see their stock fall after announcing massive CapEx due to fears of pinched profits. Simultaneously, other software stocks are penalized for not investing enough in AI. This reflects deep investor uncertainty about the timing and ROI of AI initiatives.

Hyperscalers and neoclouds cannot sustain triple-digit CapEx growth indefinitely. Jordan Nanos explains that organizations are hitting physical and organizational boundaries in mobilizing resources for gigawatt-scale sites, industrial equipment, chips, and cooling systems. Finding trillions of dollars in capital becomes increasingly difficult, meaning spending growth must inevitably decelerate even though underlying customer demand remains robust.

Despite narratives of scarcity, GPU utilization in some data centers is as low as 35-40%. This is due to colossal hoarding and double-ordering by companies terrified of being caught short in a future parabolic moment. This behavior creates a phantom demand and points to a future supply glut.