Defense tech boom goes mainstream: The Economist declares Silicon Valley's warrior spirit is back
Feb 17, 2025
Key Points
- Defense tech dealmaking in America rose more than a third over two years to nearly $40 billion, even as overall venture funding fell, signaling a decisive cultural shift in Silicon Valley toward military contracting.
- Castilian, a 2022 startup founded by three SpaceX alumni, is already test-launching hypersonic missiles with the US Air Force using cheaper automotive-grade chips instead of expensive space-grade components.
- Palantir's CTO Shyam Sankar argues the Pentagon must adopt venture's power-law model and shift to fixed-price contracts to sustain the defense tech boom, as cost-plus arrangements fundamentally misalign incentives for startup economics.
Summary
Defense Tech Goes Mainstream—Silicon Valley Swaps Libertarianism for Patriotism
The defense tech boom is no longer fringe. The Economist's reporting on the sector crystallizes what Silicon Valley insiders have watched unfold over the past five years: a fundamental shift in how venture capital and tech founders view military contracting, driven by Ukraine, China, and the simple fact that defense startups now generate enormous returns.
The scale of the shift is stark. Defense tech dealmaking in America rose more than a third over the past two years to almost $40 billion, according to PitchBook—this while venture funding as a whole fell over that period. The marquee numbers are hard to ignore: SpaceX is worth $350 billion, Palantir $250 billion (more than Lockheed Martin, Northrop Grumman, and General Dynamics combined), and Anduril is raising $2.5 billion at a $28 billion valuation. Castilian, a startup founded in 2022 by three SpaceX alumni in El Segundo, California, is already test-launching hypersonic missiles with the US Air Force.
What makes this remarkable is not the weapons themselves but the cultural inversion. As recently as two years ago, Castilian couldn't open a bank account in Silicon Valley due to stigma around weapons manufacturing. Today, even infrastructure providers like Rippling ostensibly block defense companies, yet the venture ecosystem is pouring risk capital into the sector at scale. The cold war defense hub in El Segundo is literally coming back to life—the same neighborhood that once dominated military space flight, then went dormant after the Soviet Union fell.
The playbook shift matters as much as the dollars. Anduril's trajectory offers the template. The company started in 2016 with purely non-kinetic assets: software and sensor towers (essentially cameras). It moved deliberately into drones, then autonomous weapons systems. Only after establishing credibility in those domains did it win the IVAS (Integrated Visual Augmentation System) program—a multibillion-dollar program of record for the US Army. This wasn't accident; founders pursued this sequencing deliberately to derisk the sector. Now five years into the boom cycle, Anduril and peers have sufficiently destigmatized defense tech that a host of unicorns and startups have followed. The market map is crowded enough that new entrants face real competition: unless you're working with novel technology, there are likely funded teams already building the same thing.
The economics favor scrappiness over scale. Castilian uses automotive-grade chips costing a few hundred dollars instead of expensive space-grade ones, and manufactures rocket motors itself to cut costs. SpaceX's ability to shuttle heavy payloads to low Earth orbit cheaply has made space an increasingly affordable layer of battlefield tech. The shift reflects how Ukraine has reshaped military hardware strategy: smaller weapons, notably drones powered by AI, have supplemented heavy armaments. That opens space for startups to build cleverer and cheaper versions faster than legacy contractors can move.
The venture model itself is under stress. Shyam Sankar, CTO of Palantir, argues that if the government doesn't figure out how to make the venture model work within defense, the boom won't sustain. The obstacles are real: the Pentagon allocates a meager share of spending to startups, takes a broad-brush approach to oversight, and still relies on cost-plus contracts that guarantee incumbent contractors decades of revenue. Sankar contends the Pentagon must embrace Silicon Valley's power-law dynamics—where a few big winners generate enormous returns compensating for many failures—and shift toward fixed-price contracts that allow outsiders to bid competitively. Cost-plus contracts fundamentally misalign incentives for the startup model.
The ideological swing is as significant as the financial one. For decades, the libertarian ethos that dominated Silicon Valley made defense work taboo. That's flipped. Patriotism and American military might are now in vogue in the Valley. Alex Karp, Palantir's CEO, is championing this explicitly in his forthcoming book The Technological Republic, arguing that Silicon Valley should prioritize work with Uncle Sam on military programs as a new corporate purpose—an idea that appeals to the Trump administration. The old revolving door between the Pentagon and defense contractors is extending to venture capitalists and tech leaders; the flow between DC and Sand Hill Road is accelerating.
There is a survivorship bias here: this segment is only five years old. Anduril, the trend-setter, has reached escape velocity, but the market map is now saturated enough that execution and genuine technical differentiation matter more than timing. The dollars are real, the tailwinds are sustained, and the cultural permission structure has inverted decisively. But the startup graveyard in defense may eventually be as deep as anywhere else in venture.