The rapid expansion of artificial intelligence is facing a significant physical constraint: the inability of the U.S. electrical grid to keep pace with rising energy demands. According to a recent report from Goldman Sachs, the "glacial pace" of building and permitting new power infrastructure has created a bottleneck that threatens to stall the massive financial investments currently pouring into the AI sector.

Goldman Sachs projects that global power demand from data centers will surge by 170% by 2030. This growth is colliding with a regulatory and construction environment that is struggling to provide the necessary electricity to support these facilities. While Silicon Valley firms are accustomed to rapid innovation, the report suggests that the physical reality of energy infrastructure requires long-term planning and permitting processes that are currently incompatible with the speed of AI development.

Wall Street analysts are increasingly expressing concern regarding the sustainability of the current debt-fueled AI buildout. The report highlights that energy availability—rather than just computing power or software capability—is now the primary risk factor for the trillions of dollars in projected spending. While the report focuses on the economic risks to investors, it underscores a broader tension between the tech industry’s desire for rapid scaling and the limitations of aging national utility systems.