It was another strong year for the S&P 500 in 2025, and the index continued to be led primarily by a narrow group of large technology companies expected to benefit from the rollout of artificial intelligence (“AI”). Collectively, these firms invested more than $400 billion in AI infrastructure, driven by the belief that AI would transform nearly every sector of the economy and deliver substantial productivity gains and cost savings across corporate America.

Recently, however, cracks have begun to appear in this narrative. Investors are increasingly questioning whether returns on such massive capital investments will justify the spending, raising concerns that the market may be experiencing another speculative cycle reminiscent of the 19th-century railroad boom or the dot-com bubble of the 1990s. Several large technology companies—notably Oracle and Meta—have already seen their stocks decline as markets penalize what is now viewed as excessive spending. History suggests that companies with the highest capital expenditures often underperform over time, and intense, well-funded competition in AI may ultimately limit returns for its developers, even as the technology delivers meaningful benefits to consumers and workers.