The U.S. Census Bureau's April construction spending release documents data center outlays exceeding combined federal, state, and local transportation expenditures for the first time in the reported series. Primary records from the Census Bureau Construction Spending Survey (C30) provide the underlying monthly estimates, derived from surveys of construction firms and government agencies rather than modeled projections. This shift occurs against the backdrop of the CHIPS and Science Act of 2022 (Public Law 117-167), which authorized semiconductor and advanced manufacturing incentives, and successive executive orders directing federal AI research priorities. One perspective, grounded in Office of Management and Budget historical tables on discretionary outlays, frames the change as an expected outcome of statutory limits on non-defense discretionary spending relative to private-sector returns on compute infrastructure. An alternative reading, drawn from Department of Transportation budget justifications submitted to Congress, notes that authorized highway and transit formulas remain constrained by the Highway Trust Fund balance and reauthorization cycles, producing slower nominal growth than hyperscale facility pipelines. A third vantage, reflected in National Science Foundation AI research program summaries, highlights risks that concentrated private capital may accelerate capability gaps in public-sector data governance and energy grid planning. These primary documents together illustrate divergent authorization timelines without implying convergence or superiority of either channel.