China is building AI infrastructure like America built bombers in 1942. We're not. In 1941, Roosevelt demanded something insane: 50,000 planes a year. Everyone thought he was crazy. By 1944, we were rolling a new bomber off the line every five minutes. The book Freedom's Forge by Arthur Herman shows how the United States won the manufacturing race of the 20th century by temporarily transforming the government into the world's most aggressive technology investor.

An auto industry executive named Bill Knudsen stepped up to lead this effort. "You have got to let business make money out of the process," Knudsen wrote in his diary, "or business won't work." The results were staggering. American factories produced two-thirds of all Allied military equipment.

It’s hard to imagine the government acting this way today but that’s exactly what we need! Here’s what I learned from the book and the implications for today's AI race. As we face growing competition with China in artificial intelligence, the lessons in Freedom's Forge suggest we reconsider our approach to large-scale technology development.

The Economics of Rapid Scaling

In WWII, the government funding model was simple cost-plus contracts. The government would pay all expenses plus a fixed fee of ~7-8 percent. Cost-plus contracts guaranteed capital for results without worrying about how to get paid back.

Between Pearl Harbor and V-J Day, the War Department advanced $7 billion to civilian contractors; the Navy added another $2 billion. The government bought massive scale with clear strategic intent.

When Packard Motor Car Company got a handshake deal from Knudsen to build airplane engines, they started work without a signed contract. They built machine tools, expanded their factory, and still launched their 1941 Clipper car. The government's guaranteed contracts provided the confidence that enabled private investments to flow.

This approach showed how different types of innovation funding work at different stages. While venture capital excels at backing new ventures, it's less suited for rapid industrial scaling during national emergencies. When you need to build factories, train workers, and solve engineering problems at breakneck speed, you need capital that prioritizes national objectives over financial returns.

Speed as Strategic Advantage

Beyond the funding model, there were no lengthy proposals. In 1942, Admiral Vickery reviewed Kaiser's three-page shipyard plan. His response: "Start Monday." The committed capital drove rapid action.

Vickery recognized that in strategic competition, time often matters more than perfect efficiency. Kaiser went from breaking ground to launching Liberty ships in under a year. By 1943, his Richmond shipyard could build a complete vessel in four days, not because Kaiser was a shipbuilding genius, but because guaranteed government capital let him experiment, fail, and iterate rapidly.

This historical lesson applies to today's AI competition. While venture capital provides excellent funding for AI startups, the timeline and risk tolerance of private investors may not align with the urgency of global competition.

Building Ecosystems, Not Just Companies

The wartime supply chain wasn't centrally planned. Knudsen required big contractors to develop their own supplier networks. For example, when Ford built B-24 bombers at its Willow Run plant, it had to bring hundreds of smaller companies along to help make and assemble all the different components.

This approach forced technology transfer and ecosystem development. Half a million new businesses emerged during the war years, each learning skills that would define postwar manufacturing.

The government-backed ecosystem drove massive investments in human capital. Boeing's Wichita plant employed 26,000 people, 40% women and almost all new to manufacturing. These workers learned skills that drove America's postwar economic boom.

Applying Historical Lessons to AI Competition

Today we face a technology race with similar characteristics to WWII: existential stakes, international competition, and the need for rapid capability development across multiple domains.

China understands this dynamic. They're building AI infrastructure with the same urgency America brought to bomber production in 1942. Meanwhile, the United States relies primarily on private venture capital to drive AI development. The US approach works well for innovation but may be insufficient for comprehensive strategic competition.

Consider semiconductor manufacturing, the foundation of AI capability. The most advanced chips require extreme ultraviolet (EUV) lithography machines. Only one company makes them: ASML in the Netherlands. The next generation uses high numerical aperture (high-NA) EUV that costs $380 million per machine.

A wartime-inspired approach might involve the US government providing cost-plus contracts for multiple high-NA EUV facilities. Intel, TSMC, and a new entrant would each receive guaranteed funding to build and operate them. This would create the infrastructure foundation that ensures the US keeps its strategic AI position.

Learning from History Without Repeating Mistakes

Freedom's Forge acknowledges that cost-plus contracts enabled some waste. For example, an infamous "$600 toilet seat" became shorthand for government inefficiency. But the waste was a rounding error compared to the value created. The wartime model enabled the jet engine, radar, synthetic rubber, and antibiotics industries while transforming America into a technological superpower.

The key insight isn't that government funding is always superior, but that strategic competition requires different approaches than normal market conditions. For example, the B-29 Superfortress program cost more than the Manhattan Project. No private investor would have funded it due to uncertain returns, long timelines, and high technical risk. The B-29 pioneered technologies we still use today like pressurized cabins, remote-controlled gun turrets, and advanced radar systems.

The Path Forward

Freedom's Forge teaches us that America's greatest industrial triumph combined strategic government investment with private sector execution. The arsenal of democracy was built through the partnership between government and industry.

In an era of renewed great power competition, this historical model offers practical guidance. Today's AI race needs both the entrepreneurial energy of venture capital and strategic commitment of government investment. The country that pulls this off will determine the technological leadership for the next century.