U.S. Leverages Infrastructure and Defense to Maintain AI Dominance

The United States is maintaining global AI dominance by controlling critical physical and software infrastructure, including hyperscale data centers, subsea fiber-optic cables, and the GPU market. Nvidia currently controls approximately 85% of the global GPU market and provides the CUDA software layer, which serves as the default environment for AI development. To solidify this lead, the U.S. government has established a commercial-military symbiosis, with the Department of Defense signing agreements with providers such as OpenAI, Google, Microsoft, Amazon Web Services, Oracle, and Nvidia to deploy tools on classified networks. The Pentagon has earmarked over $54 billion in its FY2027 budget for autonomous warfare and drone systems.

Economically, AI-related investments have driven significant portions of U.S. GDP and stock market gains in 2026. Treasury Secretary Scott Bessent views AI-driven productivity as a means to avoid fiscal deficits, while President Donald Trump is exploring ways for the government to become a major stakeholder in AI companies. However, some analysts warn that valuations have reached levels not seen since 1929, potentially signaling a financial bubble.

In contrast, the Government of China is pursuing a strategy of societal integration and affordability through its "AI Plus" program and the "digital silk road" to build parallel infrastructure. Chinese models from companies like Moonshot and DeepSeek currently handle nearly two-thirds of global AI use by token volume, offering high performance with fewer resources. The U.S. continues to counter these efforts through the CLOUD Act, export controls, and regulatory blocks on undersea cables.