Build it. They're coming. 

In turbulent eras, countries face three choices: attempt to restore the past, struggle to prolong the status quo, or build for the future. The AI Action Plan and related executive orders definitively call for the third.  Released on Wednesday, July 23rd, the Action Plan involves three pillars: accelerating AI innovation, building American AI infrastructure, and leading in AI diplomacy and security abroad. Related orders help implement those high-level aims by, for example, limiting the amount of federal funds available to states that enact restrictive AI regulations and easing the permitting path to constructing data centers. The pillars jointly set the foundation for building a future that involves American AI leadership.

Some folks on the Hill, however, think that the first or second options are still on the table. In alternative universes, those options make sense. In a relatively tranquil and segmented geopolitical world, one nation's decision to return to a bygone era might have little to no impact on its neighbors. In a world in which geographical borders constrained technological advances, preservation of the present might not be problematic. Some nations could opt to explore the latest tech, while others were content to maintain current systems.

A world characterized by a (mostly) interconnected economy and defined by global transfers of technology–our world–takes those options off the table. China, Russia, Saudi Arabia, the UAE, Israel, South Korea, and many other nations chasing the frontier of emerging technologies do not care if the U.S. attempts to pause or even reverse the development and deployment of AI, the use of blockchain, and the exploration of biotech. They're going to build. They're going to construct ever larger data centers, invest in new energy projects, and accelerate startups pursuing emerging technologies we haven't even heard of. 

Some Americans may not like the administration behind the AI Action Plan. They have vehemently protested actions that have tested the rule of law and departed from democratic norms. That's understandable and, in some cases, merited. But when it comes to this plan and, more generally, a call to build—to construct new infrastructure, to establish new talent pipelines, and to create new institutions—Americans need to unite. In particular, the public should support the administration's efforts to bolster Little Tech.

The AI Action Plan and related executive orders have the potential to spark a new wave of startup activity. Cheaper and more reliable energy, deeper and more widespread expertise among the workforce, and increased access to AI inputs, such as compute, are some of the key policies announced by the Trump administration. Each of these makes it easier for small and young firms to enter this burgeoning market. That’s a wise policy move today and for the foreseeable future.

Little tech is much more than a buzzword announced in an essay penned by venture capitalists Marc Andresseen and Ben Horowitz last year. The tinkerers and innovators working in garages and studios in Austin, San Francisco, and New York are our key to a brighter future. This isn't a partisan take but one grounded in robust economic analysis. 

The small firms targeted and championed by the AI Action Plan have historically accomplished three things: first, provide people with job opportunities during otherwise economically disruptive periods—new firms often play a leading role in helping individuals laid off by major firms during economic downturns find new employment opportunities; second, set our posterity up for success by putting us on the vanguard of new technology; and, third, ensure our protection by equipping our military with the technology necessary to deter our enemies. The same is true of today's small AI labs, upstarts offering blockchain-enabled products and services, and young companies pursuing biotech advances. Examination of forecasts for the near future suggests all these attributes will be essential in the coming years.

AI and related technologies will cause job displacement. In some sectors, it will be rapid and extensive. In others, it will be slow, if not undetectable. What’s certain is that many Americans of different skills and in various parts of the country will need new avenues to meaningful work. Those jobs are unlikely to come through established firms—incumbents are the ones leaning into AI and automating folks out of their jobs. Startups are the key to opening doors to those kicked to the curb. AI-first firms are on the leading edge of markets that don’t yet exist. Many of them will fail. But many of them can provide Americans with good work for a year, if not longer. 

The success of these upstarts hinges, in part, on their ability to discover new technology and bring it to market. Thankfully, that’s exactly what startups are primed to do. Whereas incumbents are known to pursue “sustaining” innovations—those that achieve gains over existing tech but stop short of undercutting the firm’s existing market advantage, startups are the chief source of disruptive innovations that give rise to new markets. It’s also worth noting that the products and services being pursued via AI and other technologies have incredible potential to improve the quality of life and reduce the cost of living. Consider how the Alpha School is using predictive AI to improve education drastically by upending the one-to-many teaching model and allowing students to develop mastery of essential skills via a personalized AI tutor. Or look at how Stanford researchers are accelerating drug development with AI and making real the possibility of better, cheaper medicine. 

One of the most consequential benefits derived from a vibrant and growing Little Tech sector comes on the national security front. Our adversaries have rushed to augment their militaries through AI-equipped weapons, ranging from autonomous drones to sophisticated disinformation campaigns. They will not stop. As discussed further below, China in particular is poised to tap into its robust AI ecosystem to establish and maintain a strategic military advantage. Our best bet for keeping pace is not to count on a handful of large tech firms to meet all of the Department of Defense’s needs. We’ve seen this strategy before, and it’s a risky one. Large firms are reliant on lengthy supply chains, long-term contracts, and often outdated processes. Little Tech, on the other hand, can adjust rapidly to develop the weapons of the future, before the DOD even realizes such weapons are possible. 

China is not only building fast—it’s building smart. The government has created a national AI development strategy, incentivized domestic AI research, and encouraged military–civil fusion to integrate technological advancements into defense capabilities. It’s investing in chip manufacturing, quantum computing, and next-generation networks. Startups in Shenzhen and Hangzhou are producing cutting-edge AI products with explicit government backing. The result is a nimble and robust innovation pipeline that ties together universities, military research, and private industry. Russia, meanwhile, is using AI to bolster its cyber warfare capabilities and propaganda machines. The Gulf states are creating sovereign AI funds and recruiting top-tier international talent to establish domestic AI capacity. Israel and South Korea are marrying their national security priorities with aggressive investment in AI R&D.

This isn’t a call for panic. It’s a call for preparation. The AI Action Plan is a blueprint for meeting the moment—but only if it is followed by bipartisan implementation, sustainable funding, and a cultural shift that celebrates building. Skepticism about AI is warranted. Cynicism about Big Tech is earned. But hesitance about building a better future—one powered by Little Tech and guided by shared democratic values—is untenable.

This is a race not for dominance, but for resilience. A future in which Americans across the country are equipped to adapt, to contribute, and to thrive will not be handed to us. It must be built. Build it. They're coming.

Kevin Frazier is the AI Innovation and Law Fellow at the University of Texas School of Law and co-host of the Scaling Laws podcast.