Winning the Next War
In national security circles, we generally acknowledge that American industrial capability brought national success during World War II. We built trucks and tanks for war in Europe, produced aircraft carriers and planes for the Pacific, and launched up to three liberty ships per day to move people and supplies around the globe.1 But today, given recent advances in artificial intelligence (AI) and robotics, the next war will be won based on the victor’s ability to power their electric machines.
Right now, the United States is not ready.
Why China seems better positioned in energy
National power in the future will be driven by electricity, and we need to write our national strategy accordingly—the country that can throw the most energy at a problem will win.
It has always been the case that countries and societies are able to grow and progress faster than their competitors based on their access to abundant energy and their ability to harness it. Now, as AI and robotics become increasingly present in nearly every aspect of life—from creating art and writing emails, to building software, manufacturing hardware, running companies, and conducting military operations—this trend is true more than ever before.
Many people understand that future growth in AI use will require more energy, but they are often just considering increased use of today’s chatbots. However, as we see open-source models improve and as leading proprietary models become commoditized, the most significant differentiator in the quality of results will come less from the size of the model and more from the amount of energy dedicated to each task.
More energy input will enable additional models operating in parallel, it will enable larger context and memory, and it will enable more tokens generally. All of this will lead to a higher quality output for each task. The United States is working to ensure the People’s Republic of China has limited access to the most efficient compute hardware and the Silicon Valley AI labs are protecting their best and brightest frontier models. But with the electricity available in China, these measures may prove irrelevant overall. And this is before we even start to seriously consider the energy use and productive output from robotics.
For China, which has been improving its electric system for years and has excess ability to generate electricity, the tie between productivity and electricity is an opportunity. In a country like the United States, which has just enough capacity for the highest current demand periods and is facing massive challenges in expanding the grid, the present situation is grim.
With excess electricity generation, China can build and power more automated factories. With more factories, they can build more products to export, build more robots for future factories, and eventually build more weapons on automated production lines. With a robust energy strategy, they are laying the foundation for an AI-driven robotic flywheel. If we don’t build a strategy to compete, we may wake up one day to discover we lost an economic cold war we did not even realize we were fighting—or wake up in the middle of a conventional hot war without the infrastructure to win.
The United States needs an energy strategy
There are several steps our nation must take now to address this existential challenge.
At the macro level, our nation’s leaders need to understand and plan for the impending ties between energy and every other national activity. We need to open the aperture on what national security means. Our national strategy must evolve beyond traditional perspectives of merely defense to include national economic and national energy strategies. With the coming omnipresence of AI, the future growth, prosperity, and security of the country will be a direct function of our ability to power it.
Without the right energy infrastructure or a plan to build it, we will fail.
(I’ve already argued that our legislative and executive branch leadership need to develop an overarching National Strategy that transcends beyond the patchwork status quo of government efforts to help our country succeed now and into the future.)
A long-term, comprehensive National Strategy supported by leaders on both ends of Pennsylvania Avenue and beyond can help guide us through this evolution of society and align our efforts in a shared vision for our future. This improved strategy should include a section for defense, a section for industrial policy, and a section for energy and infrastructure: the National Security Strategy, a National Economic Strategy, and a National Energy Strategy, respectively. The 2022 National Secuturity Strategy briefly covers some of these topics, but we need to go into more depth.
We can turn to military and business strategic minds and analytical frameworks to evaluate our competitiveness and develop a winning position for the United States. From such an analysis, we might note a strategic disadvantage in manufacturing capacity, production automation, and excess energy as compared to China.23
In energy specifically, we should use every tool and resource we have. We still need fossil fuels and would do well to continue exporting them. We should consider following the Norway model of leveraging significant fossil fuel exports to build national wealth.
We would also benefit by aggressively developing domestic energy diversity and longevity through renewable options—which are, by definition, the only options we will eventually have.
In the same vein, we need to push nuclear power by driving streamlined permitting, small-modular reactor development, and a concerted effort to win the world race for fusion. Whichever country first achieves sustainable fusion energy will rule the globe for the next few centuries, especially when combined with the energy-to-productivity translational powers of AI.
Looking to the future
We have not yet lost the modern global competition, but our competitors are strong. As we look to the future, we need to stop positioning for the next two or four years and start making strategic moves for the next 40 or 80.
Artificial intelligence is changing the source of international power from education and technological advancement to pure electricity, energy resources, and our ability to harness them. We need a true grand strategy—a National Strategy with the right industrial and energy policies—to endure this next chapter of disruption.
— You can explore my thoughts on our need for an overarching National Strategy here