What Investment Data Implies about the AI Transition (www.nber.org)

A recent study highlights a striking trend in the U.S. technology sector, where the five largest firms have significantly ramped up capital expenditures—totaling $380 billion in 2025, with projections to nearly double that in 2026. Such unprecedented spending underscores a critical turning point for these companies, as they face the risk of bankruptcy unless profits rise in tandem with investments. The researchers used a two-sector open-economy model, projecting that this surge in investment could lead to a productivity boost in the AI sector by 2.7 times, reflecting a potential transformational shift in the economy. The implications for the AI/ML community are profound, as forecasts estimate that by 2030, cumulative GDP growth driven by AI could range from 5 to 58 percentage points. This shift suggests that AI could represent a substantial portion of the economy—between 8% to 39%. Moreover, the anticipated long-term annual growth of approximately 7% comes with notable uncertainty. The study also indicates that rising risk aversion may lead to an increase in the risk-free rate and equity premiums, highlighting the financial stakes and the volatile landscape as the transition to AI gathers momentum.