Export Controls and The Art of War: What Sun Tzu Might Say About Semiconductor Policy

Welcome to The Strategic Technologist. I'm Mark, and I've spent the past 25 years in national defense on active duty, in the private sector, and as a government civilian. During that time, I transitioned from using technology systems to building them, then implementing the policies that govern them. Now as a tech policy director at a think tank, I'm interested in exploring what happens after we pass tech policies: the strategic consequences that emerge once the headlines move on.

All Views in this newsletter are my own and do not represent the views of The R Street Institute, the US Navy, or any other organization I am affiliated with.

This week, I was quoted in an article in Politico about semiconductor export controls and the position I took, that export controls of AI chips to China prioritizes short-term thinking while presenting a greater long-term strategic risk, countered the thinking of national security experts.

This got me thinking about why so many smart and qualified people disagree on this issue. The conventional wisdom treats export controls as a technique in a chess game. If we deny China certain pieces in the game, then we have a greater chance of winning. While denying your competitor pieces in a game gives you an advantage, this is not how strategic competition works in the real world.

Sun Tzu understood this dynamic 2,500 years ago and laid out a hierarchy for strategic approaches: the highest form is breaking enemy resistance without fighting, followed by attacking their plans, then disrupting their alliances, then attacking their army, and finally besieging their cities (The Art of War, Chapter 3). We can use this today to evaluate our approaches to strategic competition with China.

While export controls aren't quite as bad as besieging a city, they represent the second worst approach in this hierarchy. Our export controls attack China's technological capabilities rather than their strategic plans. This strategy forces China to develop alternatives to our technology, accelerating their efforts to create an independent AI ecosystem that operates outside Western influence.

So what would attacking China's strategic plans look like instead of focusing on their capabilities?

First, we need to correctly identify China's strategic center of gravity. It's not their AI development or semiconductor capabilities, it's their drive for technological sovereignty that allows them to emerge as a great power without Western dependence. Everything else, such as AI ecosystems, chip manufacturing, alternative tech standards, are supporting pillars of that broader strategic objective.

Export controls attack one supporting pillar while accelerating the center of gravity itself. By denying China access to our technology, we're validating their belief that technological independence is essential for national security.

A more strategic approach would target that center of gravity directly by making technological sovereignty economically and strategically counterproductive. Instead of pushing Chinese companies toward independence, we could make them so integrated into Western capital markets, talent pipelines, and technical ecosystems that sovereignty becomes prohibitively expensive and strategically isolating.

This asks a fundamental question about the nature of our strategic competition itself. Are we trying to maintain permanent technological superiority through denial, or gain strategic advantage through integration and dependence? The first approach assumes we can control the pace of Chinese technological development indefinitely, which is an impossibility in the digital age. The second recognizes that our advantage comes from being the center of global innovation networks, not from blocking others from innovating.

The path of export controls leads to a bifurcated global technology ecosystem, which is exactly what China wants and what we should be trying to prevent. If we succeed in forcing China to build parallel systems for semiconductors, AI, cloud computing, and digital payments, we'll have handed them the technological sovereignty they've been seeking while fragmenting the global systems that give us strategic advantage.

Sun Tzu would recognize this as a classic strategic error: we're choosing the direct confrontation that makes our opponent stronger rather than the indirect approach that preserves our position. The question isn't whether we can slow China down for two or three years. The question is whether we can build and maintain the kind of strategic competition we can win over the next two or three decades.

"What This Means For..."

U.S. strategic competition: We're at a decision point about how we want technology to factor into strategic competition. Export controls push us further into a confrontational model that plays to China's strengths in state-directed development. Integration strategies play to our strengths in innovation ecosystems and alliance coordination.

Policymakers: Tactical wins in export controls may actually be strategic losses if they accelerate Chinese technology independence. Success should be measured by Chinese dependence on Western systems, not by our ability to deny them specific technologies.

Tech companies: While most companies have limited influence over US-China tech policy, those that do should prioritize maintaining access to both markets where possible, rather than prematurely choosing sides in what may ultimately be a temporary policy approach.

Aspiring strategic thinkers: Strategic principles from 2,500 years ago remain remarkably relevant today. In technological competition as in war, making your opponent desperate often makes them dangerous. In this case, the strategic goal should be making technological sovereignty unattractive, not inevitable.