US presidents rarely get to choose the challenges that define their place in history. So it will also be for President Donald Trump, for all his efforts to set the agenda during his first two weeks in office.
It’s a fair bet that a century from now, Trump’s early emphasis on immigrant deportations, tariffs, Greenland, and Panama won’t be as long-remembered as whether he undermines, sustains, or increases the United States’ global standing in relation to China and its autocratic allies.
Little will be more significant in that effort than whether the United States can rule the commanding heights of technological change. It’s that question that should weigh most heavily on Trump as he considers how to manage the artificial intelligence (AI) race with China following this week’s news that the Chinese company DeepSeek has achieved AI results as good or better than some American models at lower cost and apparently without the most advanced chips.
A couple of days before Trump’s inauguration, outgoing National Security Advisor Jake Sullivan talked with Jim VandeHei and Mike Allen of Axios about the catastrophic risk in losing this contest. Their interview didn’t get the attention it deserved, so read about it here now if you haven’t already.
What VandeHei and Allen took away from the conversation with Sullivan was that “[s]taying ahead in the AI arms race makes the Manhattan Project during World War II seem tiny, and conventional national security debates small. It’s potentially existential with implications for every nation and company.”
Sullivan would be the first to concede the flaws in the comparison between the AI race and the race to a nuclear weapon. The Manhattan Project involved an array of technical problems that were frontier physics with a clear government coordinator, whereas with AI, the challenges are largely being solved in universities or commercial research labs, without the power of the US government coordinating.
That said, the point of the comparison is that the outcome of the race could have generational consequences of similar magnitude around what country or set of countries sets the rules for the future.
Distilling Sullivan’s comments, the Axios authors add: “America must quickly perfect a technology that many believe will be smarter and more capable than humans. We need to do this without decimating U.S. jobs, and inadvertently unleashing something we didn’t anticipate or prepare for. We need both to beat China on the technology and in shaping and setting global usage and monitoring of it, so bad actors don’t use it catastrophically. Oh, and it can only be done with unprecedented government-private sector collaboration—and probably difficult, but vital, cooperation with China.”
There are some Chinese advantages, underscored by DeepSeek, that the United States will find difficult to match. As Yuan Gao and Vlad Savov of Bloomberg explain, “The country has a deep pool of highly skilled software engineers, a vast domestic market and government support in the form of subsidies as well as funding for research institutes. It also has a pressing necessity to find a way to do more with fewer resources.” They could have added that China’s principal advantage is massive, unfettered data access without any of the complications of privacy concerns.
Most of all, the Chinese government and its private companies work hand in glove. This is perhaps the biggest challenge for the United States, and it is also the biggest difference between now and during the Manhattan Project.
Sullivan told Axios that unlike previous tech breakthroughs where the United States found a way to lead—atomic weapons, space travel, and the internet—AI development “sits in the hands of private companies with the power of nation-states,” VandeHei and Allen write.
What does this difference mean? To begin with, the US government will have to work more effectively with private tech companies than ever before if the country is to sustain its early AI lead and shape global regulations around it. Trump will also need his democratic allies on board. Unfortunately, many of these allies are busy at the moment hatching approaches to counter Trump’s tariff threats and, in Europe, weighing how to respond to his aspirations to gain control of Greenland, an autonomous territory of the Kingdom of Denmark.
It doesn’t take cutting-edge AI to decipher that the new administration already has daunting and far-reaching choices before it.
Frederick Kempe is president and chief executive officer of the . You can follow him on X: @FredKempe.
This edition is part of Frederick Kempe’s Inflection Points newsletter, a column of dispatches from a world in transition. To receive this newsletter throughout the week, sign up here.
Image: Deepseek app is seen in this illustration taken, January 28, 2025. REUTERS/Dado Ruvic/Illustration/File Photo
