Broken defense procurement is holding back American military readiness — and Silicon Valley wants to fix it

Summary

Pentagon procurement is broken. Silicon Valley now has the political opening to fix it.

The U.S. defense acquisition system has become so dysfunctional that it cannot execute tasks the military accomplished routinely a century ago. The comparison is instructive: in 1911, the Army tested six pistol prototypes, ran each through thousands of rounds over 48 hours, and selected the Colt 1911—a decision that held for over 70 years. By contrast, the Army's modern attempt to buy a new service handgun consumed 10 years and generated 30,350 pages of requirements without producing anything.

The problem, as Joe Lonsdale and collaborators outline in Pirate Wires, is ideological. The Pentagon purchases more goods and services than all other federal agencies combined, yet an ideology of central planning has rendered it ineffective. Each inefficiency triggers congressional shock, which prompts the bureaucracy to create more inefficiency in response—more reports, more analysis, more regulation. It's a cycle that privileges process over action.

The core pathology is requirement creep. The Navy's Constellation-class frigate offers a stark example: it was meant to be a fast, off-the-shelf purchase of the proven Fremm frigate, with Congress explicitly demanding a quick acquisition. Navy bureaucrats then reduced design commonality from 85 percent to less than 15 percent—meaning they rewrote 85 percent of the ship—by layering decade-old requirements onto the original design. The result is massive cost overruns, schedule delays, and a Navy still without a frigate in service.

Because the Pentagon is the only domestic customer for defense products, contractors face no competitive pressure to improve efficiency. Programs that would never survive open market competition can thrive if the Pentagon wants them. Superior products struggle to break through. When contracts do get awarded, losing bidders sue—a rational response to opaque processes—and litigation becomes standard.

The new Trump administration has both motive and personnel to reshape this system. Pete Hegseth, the Defense Secretary, emphasized competition and criticized the Pentagon for becoming too insular and blocking new technologies. Steven Feinberg, the nominee for Deputy Secretary of Defense, brings private equity discipline and extensive defense deal experience. Emil Michael, who was Travis Kalanick's chief lieutenant at Uber, now has influence over Pentagon research and engineering—and the administration wants that startup energy inside the building.

The specific proposals gaining traction include breaking the Pentagon's monopsony power by putting service branches into direct competition with each other for contracts, not just centralizing procurement decisions. Another is to replace Byzantine requirements with mission-based outcomes. Instead of writing hundreds of design specifications, the Pentagon would state the operational problem—say, "intercept aerial threats faster"—and let industry compete on solutions.

There's also appetite to reform the Federal Acquisition Regulation (FAR), the ironically named rulebook first issued on April 1, 1984, that has nearly doubled in length. Title 10 of the U.S. Code, which governs the armed forces, has become swollen with obsolete legal requirements that invite the lawfare that routinely derails procurement. SpaceX and Palantir both had to sue the government when contracts didn't go their way, and that pattern will persist until the legal and regulatory framework changes.

The political moment is real. VCs historically avoided defense tech because it depended on government procurement—a bet that looked too risky. That calculation shifted after Uber, Lyft, Airbnb, and SpaceX kicked down doors with faster iteration and better products. Since the Cold War, over 2,500 U.S. companies have reached unicorn status; fewer than a dozen have been in defense. The gap reflects not technology gaps but acquisition dysfunction.

Defense startups now face a brutal choice: absorb years of acquisition delays while burning $2 million per month in payroll, or abandon the market. The result is that otherwise superior technology gets shelved. The counterargument—that even complex systems like the F-35 ultimately work despite massive overruns—misses the point: the U.S. isn't losing a technology war, it's losing an efficiency war. That's fixable with regulatory pruning and cultural change. Whether the new administration moves fast enough remains the open question.