Opposition to data center construction and growing distrust of technology corporations are creating conditions for a reassessment of the pace of artificial intelligence development in the United States. Another factor could be the crisis surrounding Nvidia chief Jensen Huang, who admitted to a relationship with a woman suspected of spying for China. As Axios notes, a decline in the political influence of one of the leading advocates for accelerated AI development could give more room to those calling for restrictions on the industry.

However, slowing the construction of computing infrastructure does not in itself guarantee that AI progress will stop. Companies already have substantial computing resources, and developers in other countries may continue expanding capacity. The main opportunity that supporters of restrictions see is to use public discontent to push for broader regulation.

Nvidia and Jensen Huang Are Losing Some Political Influence

For several years, Nvidia's chief has been one of the most influential advocates for the rapid development of artificial intelligence. He has pushed back against concerns about a potential AI market bubble and criticized notions of catastrophic threats that could arise from the emergence of extremely powerful systems.

Now, however, Huang's position has become more complicated. After admitting to a relationship with a woman suspected of ties to Chinese intelligence, he may face increased scrutiny from the board of directors, U.S. lawmakers, and federal agencies.

Nvidia says Huang will remain CEO and that the scandal will not affect the company's operations. But the consequences for his political clout are harder to assess. If investigations limit his ability to publicly advocate for the continued expansion of the AI industry, this could weaken one of the most influential figures opposing a slowdown in the technology race.

The significance of the situation extends far beyond a single company. Nvidia remains the largest supplier of AI accelerators while also investing in businesses building artificial intelligence infrastructure. Questions about its leadership, business practices, or financial ties could therefore affect expansion plans across the entire industry.

Opposition to Data Centers Is Becoming a Political Factor

The crisis surrounding Huang is unfolding alongside growing resistance to the construction of new data centers. As Axios notes, electoral victories by Democratic candidates and initiatives to impose moratoriums on the construction of such facilities are increasing political pressure on technology companies.

The reasons for the discontent go beyond attitudes toward artificial intelligence itself. Large-scale data centers require substantial energy resources and put pressure on local infrastructure. As a result, projects that until recently were seen as symbols of technological progress and economic investment are increasingly becoming the subject of public and political disputes.

Restrictions could make it harder to expand computing capacity in the United States. But this would not necessarily lead to a comparable slowdown in AI development: companies can move construction to other countries, while existing infrastructure will continue to operate.

Developers Acknowledge the Risks but Are Unwilling to Stop First

The paradox of the current situation is that the leaders of the largest AI companies have themselves repeatedly warned about the potentially catastrophic consequences of developing advanced artificial intelligence.

Anthropic chief Dario Amodei, OpenAI chief Sam Altman, and SpaceX chief Elon Musk have at various times spoken about the risks associated with increasingly powerful systems and the need to control the pace of their development.

However, there is a serious contradiction between acknowledging the danger and being willing to halt one's own development efforts. No company wants to voluntarily cede ground to competitors, especially amid rivalry with China. If one developer slows down, others may continue the race and gain a technological advantage.

As a result, even companies advocating precautionary measures find themselves participating in a competition that makes a coordinated slowdown difficult.

Why a Moratorium on Data Centers Is Not Enough

Anthony Aguirre, head of the Future of Life Institute, believes that slowing data center construction would be helpful, although it would not fully solve the problem.

According to him, such restrictions are unlikely to automatically give society an extra year or two to prepare for the emergence of more powerful AI. Developers already have enormous computing resources, and AI systems themselves are increasingly helping improve technology and accelerate research.

Nevertheless, Aguirre sees a more important political opportunity in the protests against data centers. Discontent over infrastructure construction could become a starting point for discussing laws and rules that would directly restrict the development of the most advanced AI models.

He advocates the concept of “pause and pivot”—halting development and changing direction. Rather than racing to create superintelligence that could potentially replace humans, Aguirre proposes focusing on systems that enhance human capabilities.

This is a more radical position than simply calling for stronger safety measures. Some researchers and company executives believe AI development should continue, but with more robust oversight mechanisms.

China Complicates the Prospects for U.S. Restrictions

Even if the United States decides to slow data center construction and the development of advanced models, this does not mean similar measures will be adopted worldwide.

Meta and other technology companies are already considering building infrastructure outside the United States. For example, Google announced last week that it planned to invest $15 billion in Malaysia over two years.

Chinese developers may also continue working without the same restrictions that U.S. authorities could potentially introduce. This raises concerns that a U.S. slowdown would allow Chinese competitors to narrow the technological gap.

However, Aguirre believes the situation is more complex. Chinese AI companies rely to some extent on distillation from leading U.S. models and technical knowledge accumulated by U.S. developers. A slowdown in American research could therefore indirectly affect the pace of progress in China as well.

Moreover, a pause in the development of the most powerful systems could create an opportunity for direct negotiations between Washington and Beijing on coordinated restrictions. Instead of treating a technological race toward superintelligence as inevitable, the countries could discuss shared safety rules.

The Debate Is Shifting from the Pace of Development to Its Very Goal

Political pressure on the AI industry does not yet mean that the United States is preparing to halt the development of artificial intelligence. But the combination of public discontent, opposition to data center construction, and problems surrounding influential industry figures could change the conditions under which decisions about further expansion are made.

For supporters of restrictions, the main goal is not simply to gain a year or two before superintelligence emerges. They want to question the very need to create systems whose potential capabilities could exceed human control.

AI companies, by contrast, are seeking to maintain the pace of development while promising to strengthen safeguards. For now, these positions remain incompatible. But growing public resistance could turn the debate over artificial intelligence safety from a discussion within the technology industry into a fully fledged political issue—with consequences for the entire global technology race.