Summary：**Scott Bessent Warns US May Soon Dominate 80% of Global Computing Power** *Treasury Secretary Scot

referrerpolicy="no-referrer"
style="max-width:100%;height:auto;display:block;margin:0 auto;">

**Scott Bessent Warns US May Soon Dominate 80% of Global Computing Power**
*Treasury Secretary Scott Bessent says the United States could soon command 80% of the world’s compute capacity, up from roughly 60% today, positioning America for decisive AI leadership over China.*

### Introduction
In a recent briefing to congressional leaders, Treasury Secretary Scott Bessent outlined a striking projection: the U.S. share of global computing power could climb to 80 % within the next few years. Currently, American data centers, cloud providers, and high‑performance‑computing (HPC) facilities account for about 60 % of worldwide compute resources. Bessent warned that if this trajectory continues, the United States would solidify an overwhelming advantage in artificial‑intelligence (AI) research, deployment, and related economic gains—potentially outpacing China’s aggressive investments in the same arena.

### Key Developments
Bessent’s forecast rests on several converging trends. First, federal incentives enacted under the CHIPS and Science Act have spurred a wave of new semiconductor fabs and advanced packaging plants across the Midwest and Southwest. Second, major cloud operators—Amazon Web Services, Microsoft Azure, and Google Cloud—are expanding their U.S.-based hyperscale campuses at a pace that outstrips comparable builds in Asia. Third, private‑equity and venture capital flows into AI‑focused infrastructure have surged, with over $120 billion committed to U.S. data‑center projects in 2024 alone. Collectively, these moves are expected to lift the nation’s raw compute capacity from roughly 1.2 exaFLOPS today to beyond 2.0 exaFLOPS by 2027, pushing the global share toward the 80 % mark Bessent cited.

### Industry Analysis
Analysts note that raw compute share is only one facet of AI competitiveness. While the U.S. leads in chip design, software ecosystems, and