The competition in artificial intelligence between the US and China is increasingly being framed as a national security issue.

Earlier this month, US President Donald Trump announced 25% tariffs on certain AI chips, including Nvidia’s H200 processor and a competing model from Advanced Micro Devices. The decision followed a nine‑month investigation into the semiconductor sector.

The measure is being viewed as part of a broader strategy to encourage US manufacturers to produce more chips domestically and to reduce dependence on foreign suppliers, chiefly those in Taiwan.

White House officials later clarified that the tariffs will be targeted. They will not affect chips and equipment imported for use in US data centers, many of which are equipped with high-performance processors for running AI models. Commerce Secretary Howard Lutnick was also granted broad authority to issue additional exemptions.

The step followed Trump’s earlier pledge to impose duties on imported Chinese semiconductors, though the related executive order has been postponed until June 2027. The president also said he might allow Nvidia to export H200 chips to China in exchange for a share of sales. This proposal raised questions about a possible conflict with the constitutional prohibition on export taxes.

Despite mixed signals, the overall direction of White House policy makes clear that AI is increasingly viewed not merely as a commercial product but as a strategic asset.

In briefing notes, including commentary from Stephanie Roth, analysts point out that the COVID‑19 pandemic exposed the vulnerability of global supply chains, when chip shortages highlighted the world economy’s reliance on stable access to these components.

“Leadership in AI is critical for technological superiority, military capability, and economic growth,” the note states.

Wolfe Research estimates that the US remains the “clear leader” in training the largest and most powerful AI models. At the same time, China has made notable progress through state‑led, capital‑intensive policies, betting on efficiency, optimization, and rapid deployment of “good‑enough” models despite limited access to cutting‑edge hardware.

In other words, the AI race is looking less like a contest among startups and more like a long strategic game between nations.