From Innovation Ecosystem to Industrial Base

Introduction

America’s defense technology boom is real. Venture-backed firms building in artificial intelligence (AI), autonomy, space, and advanced manufacturing are winning larger contracts and deploying operational capabilities closer to the frontlines. Organizations like the Defense Innovation Unit—a Pentagon office that helps bring commercial technology into military use—and more flexible contracting pathways have helped narrow the long-discussed valley of death between prototype and production.

But the sector’s strongest companies are hitting a second, more formidable barrier: the transition from production to programming. This is the leap from annual, discretionary buys in the tens of millions of dollars to multiyear, institutionalized funding of $100 million–plus through programs of record. It is the difference between selling technology to the Pentagon and having that technology being built into the military’s long-term plans, budgets, doctrines, and operations.

The Department of Defense’s recent push toward developing a Warfighting Acquisition System correctly identifies the problem, but it must now translate production into long-term budgets and acquisition decisions. Absent changes in how acquisition leaders and Congress manage tradeoffs and enable space for new entrants, this second valley of death will stall the sector, producing a generation of promising defense tech companies without the scale or durability needed to form the new defense industrial base (DIB) the nation urgently needs.

Key Takeaways

• The twin valleys: The first valley is going from prototype to production; the second is going from production to programming.
• Vendor versus prime: Firms cannot become enduring industrial players on contract renewals alone; they require budget-line durability.
• The capital risk: If successful firms cannot convert into high-growth “primes,” venture capital will pull back, stifling the formation of the next generation of defense start-ups.
• Tradeoffs are mandatory: Institutionalizing innovation requires shrinking legacy systems to fund new, superior performers.

The Innovation Ecosystem: Moving the Needle

The Department of Defense is buying more relevant commercial technology than it did a decade ago, and nontraditional firms are no longer confined to the periphery. Venture-backed companies now compete for meaningful work across autonomy, software, and space—areas once dominated by incumbents. That shift is showing up in the data: Defense start-ups captured 1.3 percent of Pentagon contract obligations in 2025, up from 0.6 percent a year earlier. While the absolute share remains modest, the trajectory is clear.

Capital has followed this shift. Firms like Anduril, Saronic, and Palantir are now competing in sectors once assumed to belong solely to traditional primes. This progress matters because modern conflict increasingly rewards software speed, autonomous mass, and rapid production cycles. While the United States will still depend on legacy platforms like aircraft carriers and submarines, deterrence and wartime endurance require the cheaper, scalable, and software-defined systems that newer firms are best positioned to provide.

The United States has built a genuine defense innovation ecosystem. But innovation alone does not create deterrence, sustain war, or build industrial power—acquisition does. For start-ups and investors, early wins can mask how fragile scale remains if discretionary funding doesn’t give way to formal programming. It is here, at the threshold of institutional adoption, that many firms hit the next barrier.

The Second Valley of Death

Many defense technology firms can now reach production. Far fewer can cross the second valley of death: the move from short-term production success into programmed procurement, sustainment, and force structure.

A company may deploy systems, win follow-on contracts, and generate meaningful revenue. Yet many remain inside the innovation ecosystem—pilots, prototypes, annual buys, and contract ceilings—rather than entering the acquisition ecosystem of programs of record, procurement lines, sustainment accounts, and multiyear demand written into the Program Objective Memorandum (POM), the Pentagon’s five-year budget plan that determines what capabilities it intends to fund and scale.

Successfully bridging that gap determines whether a company becomes a durable defense enterprise or remains a promising vendor. Crossing this valley is difficult for three structural reasons.

• Fragmented ownership: Operational commands can validate a capability and fund limited use, but they rarely control the long-term budget lines required for scale. Program offices control the money but are often incentivized to execute existing portfolios rather than absorb disruptive entrants.
• The zero-sum budget: New capabilities rarely receive “new” money; they require shrinking or terminating an existing effort. Newer firms compete not just on merit, but against sunk costs, legacy constituencies, and contractors already embedded in the budget.
• The veto gap: As decisions move toward service headquarters and Congress, more stakeholders gain veto power while fewer own the operational outcome. Leaders default toward what is familiar, politically defensible, and already programmed.
  

Implications: A Theoretical Arsenal

Without a bridge to programming, much of today’s defense tech sector remains a theoretical arsenal: firms with real products and growing valuations but without the purchase scale or long-term certainty required to become enduring industrial powers. This leads to three systemic risks.

• Zombie primes: We risk creating a class of firms too large to fail or be acquired, yet too small to generate the scale required to transform the market. They survive on renewals but never achieve the weight of a true defense prime. Without scale and certainty, firms hesitate to hire workers, build factories, and deepen supply chains.
• Compressed returns and cooling capital: If successful firms cannot convert into high-growth primes, venture returns will compress toward traditional defense multiples. If the path to the POM remains blocked, the private capital that fuels this ecosystem will inevitably pull back.
• A force without mass: For the Joint Force, the stakes are existential. If companies building autonomous systems and scalable manufacturing cannot grow, the military may preserve legacy capacity while failing to generate the next layer of combat power needed for competition and any potential Indo-Pacific conflict.
  

Recommendations

Many promising reforms still exist more on paper than in portfolios. To bridge the second valley, the Department of Defense and Congress should adopt five shifts.

• Assign ownership early: Once a capability reaches production, service acquisition executives should designate a portfolio owner responsible for its transition to a program of record.
• Bridge funding with exit criteria: Congress should expand transition funds to move proven capabilities from production to programming, with mandatory “scale, defer, or sunset” decisions.
• Force POM reviews for proven systems: Operational success in the field should trigger a formal budget review, preventing systems from languishing while waiting for a budget cycle to open.
• Reward divestment and substitution: Reform must be measured by real tradeoffs—curtailing underperforming legacy systems to fund new, stronger capabilities.
• Modernize budget pathways: Expand flexible funding categories for software, autonomy, and AI-enabled services that do not fit neatly inside legacy 1960s-era “hardware” budget buckets.
  

The question is no longer whether the defense tech sector can deliver the innovative capabilities the warfighter needs. It is whether U.S. institutions can convert that innovation into enduring industrial power before the strategic window closes.

About the Author

Brian Katz works at the intersection of defense technology, venture capital, and national security policy. He is currently a mission director at Vannevar, a venture-backed defense technology company. He previously served as a policy advisor in the Office of the Under Secretary of Defense for Intelligence and Security and in the Office of the Under Secretary of Defense for Research and Engineering, led technology initiatives as a fellow at the Center for Strategic and International Studies, and spent a decade in the U.S. intelligence community. He is a U.S. Navy Reserve intelligence officer currently serving with the Defense Innovation Unit. Katz is a graduate of the Stanford University Graduate School of Business, Johns Hopkins School of Advanced International Studies, and Duke University. He is a proud former CNAS Shawn Brimley Next Generation National Security Leaders fellow. The views expressed are his own and do not represent those of the Department of Defense, the U.S. government, or Vannevar.

About the New American Defense Industrial Base Series

This essay series, The New American Defense Industrial Base, features expert practitioners with experience in government, industry, and finance writing on the most pressing challenges in defense acquisition today. For more in this series, click here. The DIB series is made possible by general support to the CNAS Defense Program and corporate support for the series.

About the Center for a New American Security

As a research and policy institution committed to the highest standards of organizational, intellectual, and personal integrity, CNAS maintains strict intellectual independence and sole editorial direction and control over its ideas, projects, publications, events, and other research activities. CNAS does not take institutional positions on policy issues, and the content of CNAS publications reflects the views of their authors alone. In keeping with its mission and values, CNAS does not engage in lobbying activity and complies fully with all applicable federal, state, and local laws. CNAS will not engage in any representational activities or advocacy on behalf of any entities or interests and, to the extent that the Center accepts funding from non-U.S. sources, its activities will be limited to bona fide scholastic, academic, and research-related activities, consistent with applicable federal law. The Center publicly acknowledges on its website annually all donors who contribute.