Ben Horowitz & Marc Andreessen: Why Silicon Valley Turned Against Defense (And How We're Fixing It)
Silicon Valley's historical alliance with US defense fractured over 15 years of hostility, but is now being rebuilt under the "American Dynamism" thesis, driven by the geopolitical realities of Ukraine and competition with China. This shift prioritizes building in the physical world—defense, energy, manufacturing—by leveraging software, AI, and autonomy to innovate faster than legacy, five-year-plan-based incumbents and win by doubling down on America's chaotic creativity.
The Great Decoupling: Silicon Valley’s Break with Washington
There was a time, from before World War II through the 1990s, when the alliance between Silicon Valley and the United States’ national mission was a given. The relationship was deeply integrated, with tech pioneers like William Shockley and Frederick Terman encouraging students to build companies supporting the war effort. This partnership was considered natural, with companies like Netscape viewing the Pentagon as a key customer and its founders holding top-secret clearances.
However, starting in the 2000s, this alliance began to crumble, escalating into outright hostility over 15-20 years. This shift was fueled by a combination of factors:
- A post-Vietnam and post-Iraq skepticism towards the military, where working on weapons was seen as inherently immoral.
- The broader national political drama, which transformed policy disagreements into moral crusades.
- A cultural disconnect, where tech workers in Silicon Valley had little to no daily interaction with military personnel, unlike in Washington D.C.
This culminated in the Google Maven project revolt, described as a “watershed moment.” An internal uprising from the “international AI community” forced Google to pull out of a Pentagon contract for AI in drone technology. This event crystallized the deep-seated animosity and forced people to take sides. The period was characterized by emotional outrage replacing intellectual curiosity. An anecdote was shared about a tech CEO who, facing employee protests for working with ICE, invited a former head of border security to explain the complexities of the mission. The entire company was invited; only three people showed up.
The Turning Point: “It’s Time to Build”
Several key events forced a dramatic re-evaluation of Silicon Valley’s role and priorities, shifting the cultural zeitgeist away from the “Harvard dorm room” ethos of the 2010s.
First, Marc Andreessen’s 2020 essay “It’s Time to Build,” prompted by the COVID-19 pandemic, highlighted America’s shocking inability to produce physical goods like PPE. It was a wake-up call that the nation had lost its capacity to make things, having focused almost exclusively on the application layer of software.
Second, the war in Ukraine provided a stark lesson in modern warfare. It revealed that the battlefield now iterates in days, not decades. The conflict demonstrated the vulnerability of large, expensive, “exquisite” platforms, which could be destroyed by cheap drones. This invalidated the slow, Soviet-style five-year procurement plans that legacy defense contractors and the Pentagon still operated on, proving them obsolete in the face of dynamic, fast-evolving threats.
This new reality created a fertile ground for the “American Dynamism” thesis: a return to funding and building companies critical to the national interest in sectors like defense, aerospace, energy, manufacturing, and public safety.
The American Dynamism Thesis: A New Investment Framework
The core idea of American Dynamism is to leverage Silicon Valley’s strengths—advanced software, AI, and rapid iteration—to solve the nation’s hardest physical-world problems. This makes previously difficult investment areas viable again.
- Hardware Reimagined: The thesis isn’t about traditional, capital-intensive hardware manufacturing. Instead, it focuses on using off-the-shelf commodity hardware as a platform for advanced software, computer vision, and autonomy. This approach, seen in the early work of companies like Anduril and Flock Safety, dramatically lowers costs and accelerates development.
- Deconstructing the Stack: In aerospace, the “Space 1.0” model was vertical integration, where companies like SpaceX built everything in-house. The “Space 2.0” approach, funded by American Dynamism, is about unbundling this stack. Companies like Apex Space focus on producing satellite buses quickly and efficiently, while Northwood Space tackles the bottleneck of ground station communication, allowing others to innovate on payloads without rebuilding the entire infrastructure.
This renewed focus has been met with a surge of downstream capital, solving the “messy middle” funding gap that previously plagued hard tech companies. It has also unleashed a new generation of founders, many of whom are graduates of SpaceX and Anduril, who are applying the manufacturing and engineering methodologies they learned to new domains.
Winning by Being “More Like Us”
The central philosophical debate is whether America should win the great power competition by becoming more like China—centralized, state-directed, and planned—or by doubling down on its own inherent strengths. The argument is that America won the first Cold War by being “more like us”: unleashing the chaotic, messy, but incredibly innovative power of free-market capitalism and individual creativity.
The capitalist approach is messy, but you have dynamism on your side… What if we lean even harder into innovation and even harder into creativity and even harder into entrepreneurship? Isn’t there so much more that we could do if we if we if we got to be more like us?
This stands in stark contrast to the rigid, top-down five-year plans that originated with Joseph Stalin and, ironically, still persist in the Pentagon’s procurement process and in corporate strategy. Such systems are too slow and fail to adapt to a dynamic world, especially a modern battlefield that changes daily. The future lies in flexibility, rapid iteration, and unleashing entrepreneurial talent.
The Future of Manufacturing and Defense
The goal is not to bring back the manufacturing jobs of the past but to leapfrog to the industries of the future. This means building advanced, automated factories to produce sophisticated products like electric bikes, autonomous vehicles, and robots. These facilities will create new, higher-skilled, and more engaging jobs.
In defense, the focus is on attritable systems: large quantities of low-cost, intelligent, and autonomous systems (like drone swarms) that change the economics of war. The US must be able to manufacture these systems at scale to compete with adversaries who have spent years building up their industrial base.
Investment Frontiers
Looking ahead, the American Dynamism thesis is “shifting left” in the supply chain to focus on foundational elements:
- Critical Minerals and Mining: Applying advanced software and AI to improve the efficiency and output of mining for essential resources.
- Components and “Dumb Parts”: Investing in the automated production of motors, components, and other hardware needed for a “war of mass.”
- Energy: Recognizing that an insatiable thirst for power—for AI compute, electrification, and national security—creates massive opportunities in generation (e.g., Radiant Nuclear), transmission, and storage.
- Offensive Space: A specific request for startups focused on protecting America’s critical infrastructure in low Earth orbit, which is the backbone of modern communication and defense.