Washington has developed a peculiar idea about how to “win” the global AI race: don’t run it. Instead of out-innovating China, many policymakers now seem convinced the winning move is to slow China down through export controls, sanctions, and restrictions. The impulse is understandable. The conclusion is wrong.
If the United States wants to stay ahead in AI, it must do what it has always done best: push the technological frontier forward, build at scale, and compete ferociously. The countries that win this era will not be the ones that restrict the most – they will be the ones that build the most.
America’s technology industry is the country’s most valuable strategic asset. But it would be naïve to think that dominance is guaranteed. In 2010, ExxonMobil topped the global market-cap rankings; only two of the top five companies were tech firms. Today in 2025, Nvidia sits at No. 1 and every one of the top five companies is a technology giant. Meanwhile, Intel, the indispensable chipmaker of the late 20th century, has struggled for years as competitors outrun it. Technology leadership is not a birthright. It must be earned, continuously.
History offers plenty of reminders. After the triumph of the Apollo program, few would have predicted that NASA would one day depend on Russia to ferry astronauts to the International Space Station, a monopoly that lasted from 2011 to 2020. It took SpaceX, through relentless innovation and reusability breakthroughs, to restore American launch capability. And crucially, NASA fostered competition by funding multiple firms rather than anointing a single national champion.
The same dynamic played out on high-speed rail. Japan opened the world’s first Shinkansen line in 1964. Europe followed more than a decade later. China, virtually absent from the high-speed rail landscape at the start of the 21st century, built the world’s largest network in unprecedented speed and at significantly lower cost. Being first offers an advantage, but not a permanent one.
Chip manufacturing is perhaps the most technically complex form of production humanity has ever attempted. The Schumpeterian rents from leading the industry, temporary profits from innovation before competitors catch up, are, by definition, temporary. Yet Washington increasingly behaves as if export controls alone can freeze global competition in place.
Yes, sanctions have worked in some cases. Russia cannot acquire advanced semiconductor orders from TSMC; its military capabilities have been materially constrained. A decade of U.S. export restrictions has slowed China’s progress in leading-edge chips. But the harder Washington tries to deny China access, the more Beijing doubles down on its effort to build indigenous capabilities. China’s acceleration in open-source AI models is partly a direct consequence of these restrictions.
Export controls make sense for genuinely sensitive technologies. But national security also requires staying ahead – and staying present. America cannot lead if its firms are stuck navigating ever-shifting regulatory landmines while competitors move fast and scale globally. Equally important – the U.S. cannot lead if its technology is absent from the global marketplace. Leadership depends on reach, relevance, and a thriving ecosystem of developers. When U.S. companies are forced to retreat from markets, that ecosystem begins to fracture – and competitors rush to fill the void.
The strategic priority is not merely controlling who buys these systems. It is ensuring the U.S. stays multiple generations ahead in designing, manufacturing and deploying them. That requires clarity and predictable policy. It means protecting the most critical chokepoints, like EUV lithography systems, through close coordination with allies. It requires expanding domestic capacity in advanced packaging and system integration. But it also requires global market access, because the U.S. advantage is not just in its chips – it’s in its software, its standards, and its 6-million-strong developer ecosystem. If American platforms are absent abroad, alternative ecosystems will rise. And once developers migrate, consider them gone.
The prevailing narrative in Washington is fixated on “stopping China.” That mindset blinds America to the real opportunities: accessing vast markets, shaping global standards, and building a democratic technology stack for the world. As more nations pursue sovereign AI, they will choose the platforms they can access – and the ones that remain globally relevant. Restricting U.S. technology too broadly simply invites others to define the norms and infrastructure of the next era.
By narrowing its commercial reach, the U.S. risks ceding influence to China, whose model is explicitly outward-facing. Beijing wants its AI systems embedded across the Global South, just as it did with 5G infrastructure. Xi Jinping talks openly about offering AI as a “public good for the international community.” Too often, America counters with a vision that sounds like “technology for ourselves.” A fortress economy is not a strategy for leadership. It is a strategy for decline.
Innovation thrives on openness, scale, and competition. AI is the next essential infrastructure, and every country will build it. The question is whether they will build it on American technology or an alternative that emerged because American firms were prevented from competing.
If the U.S. wants to win the AI race, it should stop trying just to slow the race down and start running faster and further. It must build more, design more, export more, and empower American firms to compete everywhere. But above all, it must ensure that the next wave of AI infrastructure is built with U.S. technology at its core. After all, this race is not a sprint – it is an Iron Man.
Mario Ottero is the Emergent Technology Policy Analyst at Americans For Prosperity, previously he worked as policy fellow at the Information Technology and Innovation Foundation (ITIF), where he worked on the intersection of innovation, trade, deregulation, and tech. Also, he worked at Cato Institute with the Human Progress team. He holds a Master of Science in Public Policy and Management from Carnegie Mellon University and is a Fulbright-CFI scholar.