DR. NADIA SCHADLOW is a Senior Fellow at the Hudson Institute.

The recently concluded U.S.–China summit highlighted that Beijing continues to double down on an economic model that is focused on dominating the industries it believes are strategically important. As U.S. officials return home and consider proposals for a board of trade and board of investment, which according to the White House could result in “increase[ed] Chinese investment into [U.S.] industries,” they should keep one principle in mind: Chinese capital and firms should remain outside of America’s advanced manufacturing buildout.

Modern manufacturing is no longer about building more factories. The emerging industrial ecosystem will fuse materials, chemicals, hardware, software, robotics, AI, and vast streams of data. Tomorrow’s factories will not merely produce goods. They will generate information, shape supply chains, and help determine future military and economic advantages. They are as much data systems as production systems.

The national security stakes are clear. Future military strength will depend heavily on commercial manufacturing. Many Department of War reforms center on producing weapon systems at scale through commercial industry. As Secretary of War Pete Hegseth has put it, the Department will “prioritize the purchase of industry-driven solutions, commercial solutions first.” Army leaders seeking to acquire tens of thousands of drones are emphasizing modular “plug-and-play” systems that can be rapidly upgraded as battlefield needs evolve. This shift is occurring across all the services. As one Army official explained, “if whatever they’re building isn’t modular with other industry partners … then I’m going to go with another industry partner.”

This trend makes the integrity of America’s industrial base even more important. Companies that build robotics, autonomous platforms, industrial software, and precision components may not have traditionally been viewed as defense firms, but many now sit at the core of the national security ecosystem.

Another national security concern is that modern factories generate enormous amounts of data about machine performance, materials behavior, and supply chains. The National Institute of Standards and Technology notes that manufacturing facilities produce vast and expanding streams of data every day. For instance, a major concept in advanced manufacturing is the “digital twin,” which is a virtual, data-driven model of a factory, machine, or production system that continuously updates based on real-world operational data. In the wrong hands, this data is industrial targeting information.

The track record for granting China access to sensitive sectors of the American economy is decidedly bad. China has repeatedly used commercial relationships, various types of business partnerships, predatory hiring, supply-chain access, and embedded corporate relationships to acquire technology and industrial know-how.

That pattern is well documented. In 2018, the U.S. Department of Justice secured a conviction against Sinovel Wind Group for stealing trade secrets from American Superconductor Corporation, including critical wind turbine source code that helped accelerate China’s wind industry while damaging the American firm. A year later, the DOJ alleged that Huawei used commercial partnerships and employee recruitment to obtain sensitive U.S. robotics and telecommunications technology. Similar allegations later emerged involving Micron Technology, GE Aerospace, and GE Power, where U.S. officials accused Chinese-linked actors of targeting advanced semiconductor and aerospace technologies through insider access, recruitment, and theft of trade secrets.

U.S. policymakers recognize these risks and have taken steps to mitigate them. The first Trump administration expanded national-security reviews of Chinese acquisitions amid concerns that Beijing was using investment to gain access to sensitive technologies and critical infrastructure. In 2018, Congress passed the Foreign Investment Risk Review Modernization Act (FIRRMA) because it understood that foreign ownership, joint ventures, and access to sensitive data can all create national-security risks. (Despite these more rigorous reviews, outcomes remain mixed, with China still acquiring U.S. industrial base companies.)

Chinese capital could also end up crowding out American investment. The Trump administration and Congress have spent significant time and political capital trying to channel U.S. government and private funding into manufacturing and industrial capacity. The objective is understandable, but it is not easy. Traditionally, because manufacturing generates lower margins than sectors such as software, it has been harder to attract private investment. Recent efforts to address this difficulty have ranged from the Department of War’s Office of Strategic Capital initiative to mobilize some $200 billion in financing to programs across the Departments of Commerce and Energy. These initiatives also aim to attract private capital, such as JPMorgan’s $1.5 trillion “Security Resilience” initiative to support “key sectors from critical minerals to frontier technologies.”

China, however, has demonstrated a willingness to tolerate lower returns in pursuit of strategic objectives. According to the Rhodium Group, Beijing has increasingly inserted “non-market considerations” into investment decisions, prioritizing strategic goals at the expense of profitability. That creates an uneven playing field for American capital, which still largely operates under market expectations for returns and efficiency. Competing against a state-backed system willing to absorb losses could make it more difficult to sustain private investment in U.S. manufacturing.

China knows advanced manufacturing will shape the next era of competition, and it has spent over a decade determined to dominate it. Today, China is the world’s largest market for industrial robots. Its most recent five year plan prioritizes advanced manufacturing as a way to increase productivity, calling for “new quality productive forces”—that is, productivity driven by robots, AI, and other smart manufacturing. Furthermore, its 2025 AI + Manufacturing initiative aims to integrate intelligent agents into production lines and supply chains as part of an overall integration of AI into 90% of the Chinese economy by 2030. China sees advanced manufacturing as a strategic asset, not just a commercial sector.

This does not mean America should wall itself off from foreign investment. The United States should deepen manufacturing ties with allies. South Korea recently pledged $150 billion to help modernize American shipbuilding. Japanese, Korean, European, and other allied firms can help rebuild critical supply chains. But China is different. It is the United States’s chief strategic competitor, a state-directed industrial rival, and a government that has repeatedly used commercial channels to advance its national aims.

Some might argue that allowing Chinese firms to build and operate factories in the United States could generate valuable spillovers for American workers and companies. For decades, China benefited from foreign firms that brought production expertise and other know-how into its economy. Couldn’t the United States do the same with leading Chinese manufacturers?

The purported benefits, however, would not likely outweigh the costs. Many Chinese companies would operate with lower overhead due to government subsidies, allowing them to sell their products at artificially low prices. Tariffs on Chinese imports allow us to level the playing field, so long as Chinese firms are kept out. The ultimate issue is that any market that includes China is neither free nor fair, and providing additional vectors into sensitive and strategic parts of the U.S. economy would be shortsighted.

Requiring Chinese companies in the United States to form joint ventures (JVs) with American companies to mitigate vulnerabilities and acquire technology could be counterproductive for similar reasons. Those who believe that the United States should or even could duplicate Beijing’s forced-JV strategy forget that the U.S. government is not the Chinese government. Unlike the Chinese government, American laws and courts enable businesses operating within U.S. jurisdictions to protect their proprietary knowledge from unauthorized transfers. Chinese companies, like all foreign companies, would be able to use such protections. As a result, American companies could simply be exposing themselves to deeper Chinese infiltration instead of gaining a true technological edge. Moreover, further tethering U.S. business operations to Chinese entities could exacerbate the industrial, technological, and economic dependencies that the Trump administration has been trying to unravel.

Others might assert that excluding China is merely a shield for uncompetitive American manufacturers and that they can only get stronger if forced to compete with Chinese firms. But American manufactures are not shielded from all competition. They already compete fiercely with companies from Japan, South Korea, and Germany, which have formidable manufacturing cultures and capabilities. In addition, as noted, competition with China is not on a level playing field.

The United States can’t afford to lose its edge at the increasingly important cross section of data, technology, and advanced manufacturing, especially since it is foundational to so many other competitive arenas, from defense to energy to AI. Rebuilding American manufacturing is urgent. But it should not involve inviting America’s principal rival into the systems on which future military and economic power will depend.

DARYL WIELAND is Senior Director of Partnerships at Palantir and an Adjunct Professor of Finance at George Mason University.

On February 24, 2022, Russia launched the largest land invasion in Europe since World War II. The prevailing consensus expected Ukraine to fold in seventy-two hours. Russian generals were killed at a rate suggesting their communications were compromised from the opening days. The Moskva, flagship of the Black Sea Fleet, was sunk by Ukrainian Neptune missiles operating on targeting intelligence that Ukraine likely didn’t develop alone.

The Ukrainians stood and fought. This was possible because of trust deposits made consistently and mostly invisibly in the eight years before Russia’s invasion.

After the Maidan revolution in 2014, Western intelligence services began a sustained program of advising, training, and critically, sharing with their Ukrainian counterparts. They developed sources, methods, assessments, and institutional intimacy that comes from working through hard problems together over time. Western advisors embedded with Ukrainian units and ran joint exercises. They shared intelligence as an operating norm. The trust account was funded piecemeal, slowly but surely. The dividend was paid in the speed and precision of Ukraine’s early defense. By the time the first Russian columns crossed the border, the trust already existed.

That is the model for a true partnership. That is the bedrock it rests upon.

What Trust is Made Of

In their book The Trusted Advisor, Maister, Green, and Galford argue that trust is a combination of credibility, reliability, and intimacy. All three are undermined when people focus too much on themselves instead of on their potential partners. They were writing about client relationships, but the framework maps cleanly onto every partnership I’ve seen succeed or fail.

Credibility is whether I believe you know what you’re talking about. Reliability is whether you’ve done what you said you’d do, repeatedly, over time. Intimacy is whether I believe you can hold sensitive information: my fears, my failures, and my strategic uncertainties without weaponizing them. And self-orientation is the degree to which everything you do is ultimately about you.

Self-orientation is where most partnerships die.

Every day I see “partners” who are credible and reliable but have high self-orientation. The advice that conveniently aligns with their product roadmap. The consultant who identifies a problem that requires six more months of consulting to resolve. Self-orientation is the sand in the gearbox of every working relationship, and it grinds the teeth of the gears as the stakes get higher.

Character is Revealed, Not Trained

The Prussian concept of Auftragstaktik is built on the premise that trust can substitute for control. Give a subordinate the commander’s intent, not the instruction. Trust him to use his judgment and even improvise when circumstances change. But a self-oriented subordinate given mission-type autonomy is not an asset; rather he is a liability who will optimize for his own narrow goals and call it “professional judgment.”

The Western advisory effort in Ukraine functioned on the same logic. Advisors built the institutional capacity for Ukrainians to decide faster and better. The advisor’s job was to help the Ukrainians learn to use that capacity themselves instead of relying on Western decision-making. By February 2022, they had.

The Ukrainians demonstrated the character and competence to persevere. Both character and competence are necessary, but they are not of equal value. An incompetent partner with high character can be developed. An extraordinarily competent partner with low character is a trap.

It takes more than a kickoff meeting to establish trust. It tends to be established in uncomfortable moments. For example, finishing up the deliverable before midnight because the deadline mattered or telling the difficult truth when a comfortable omission was available. One failure can cancel multiple exemplary performances in the psychological ledger of trust. If I cannot predict your behavior under pressure, I will stop depending on you when the stakes matter.

Character is what makes intimacy possible, because you can’t be strategically vulnerable with someone whose character you can’t trust.

The Vulnerability Threshold

I share with account teams all the time that if you partner with someone who has the same capabilities as you, acts like you, and has the same experience as you, it is unlikely to end well. You will trip all over each other as you both have the same strengths. You will also continue to have capability gaps. Rather, a thorough make/buy/ally analysis where you decide what to build yourself, what to purchase, and what to pursue through partnership is critical in defining the partner who fills your capability gaps. This probably means the partner has a different set of experiences. Likely they approach the problem from a different point of view. This will be uncomfortable and confusing at times. It can also get you to places that you can’t get to on your own.

Successful partnerships require vulnerability. You need to allow a partner to see the constraints you are operating under, the strategic bets that might not pay off, and the fears you carry into every major decision. Most partnerships mistake lavish dinners or offsite team building events for intimacy, but these social experiences can’t be weaponized because there is nothing at risk.

Strategic intimacy is different. It is letting a partner see the thing you’re most afraid won’t work. The first move is always unilateral. You share before you know if it will be reciprocated.

This is exactly what happened with Ukraine. Early in the post-2014 period, Western partners had to make a judgment call about whether they should share sensitive collection methods with a country whose own institutions were still being reformed, knowing that a single compromise could put sources at risk across an entire theater. They shared anyway, deliberately. The Ukrainians reciprocated by protecting what they received, and over time both sides raised the threshold of what they were willing to share. By the time it mattered, intelligence could flow at the speed of operations rather than the speed of bureaucracy.

It takes courage to be vulnerable. It takes low self-orientation to absorb a partner’s vulnerability without exploiting it. Credibility and reliability can produce a functional working relationship. Only intimacy allows a partner to know what you need before you ask.

The Defense Industrial Base: Where Trust Gets Tested

This lesson is as applicable in business as on the battlefield.

The relationship between established defense primes and emerging defense technology companies is, in theory, a natural partnership. Primes have scale, contract vehicles, and knowledge of the Pentagon’s procurement labyrinth. New entrants have speed, innovation, and the willingness to build capabilities the primes can’t or won’t build themselves. Each has what the other needs and yet in practice, there is a high failure rate.

The Valley of Death is real for startups. In the typical prime partnership model, the focus is on securing a single piece of IP or component rather than supporting the company that created it, giving primes little reason to invest in the startup’s broader success.

The pattern is familiar. Through tech scouting, a prime identifies a startup with a compelling capability. There is enthusiasm, an MOU, maybe even a joint demo. Then the startup discovers that the prime’s integration timeline is measured in years. The prime discovers the startup has fourteen employees and no experience in managing a subcontract data requirements list. The MOU gathers dust. The capability gap persists. The prime’s internal antibodies re-emerge arguing that if their engineers were just given the money, they could build it themselves.

The problem is almost always self-orientation. The prime treats the startup as a subcontractor to be managed rather than a partner to be developed. The startup treats the prime as a piggy bank rather than an institution with legitimate constraints.

To do this differently you need to be willing to be vulnerable. This means following through with small commitments such as quickly responding to email, making the promised introduction, and delivering the test environment on time. One of the hardest steps is the honest disclosure of constraints. For a startup, that means transparency about burn rate and the features that aren’t ready yet. For a prime, it means candor about internal politics with the engineering team that affect the partnership decision.

The prime that mentors a startup through its first classified integration by sharing lessons learned, making introductions, providing honest feedback is making the same kind of trust deposits the Western advisors made in Ukraine. When a crisis emerges and the Pentagon needs a capability fielded in months, the prime with a network of partners who are already integrated will deploy while competitors are still negotiating teaming agreements.

Bedrock Takes Time

The Ukrainians built their intelligence partnerships over years of patient and often unglamorous work. They participated in joint exercises that no one heard about. They made institutional reforms that took years to show results. They were willing to be transparent about what they didn’t know. When the crisis came, the bedrock held firm.

Trust is an operational capability. It is built in small deposits and spent in large withdrawals. The question is whether you are willing to start making deposits now before you know what crisis will demand the withdrawal.

GREG LITTLE is a senior counselor at Palantir Technologies.

AARON JAFFE is a deployment strategist at Palantir Technologies.

Operation Epic Fury exposed something most people across the Department of War (DoW) already understand at some level, but rarely say directly: a large portion of what we call “business systems” do not work in the conditions of a real fight.

For years, we’ve operated under a convenient mental model that separates the enterprise into two distinct worlds. On one side are warfighting systems: the ones that find, fix, track, target, engage, and assess. On the other are business systems: the machinery that keeps the institution running (think sustainment, supply chain, finance, human resources, procurement, and acquisition). It’s a clean division; one side wins wars, the other runs the business. But that distinction was always more administrative than real, and Epic Fury made it impossible to ignore.

Business systems are warfighting systems.

When the War Machine Breaks Down

Because Epic Fury was distributed across domains and coordinated through multiple commands, systems were forced to function as one war machine. Decisions depended on data that had to be current, trusted, and immediately usable. There was no pause between operating and sustaining, no buffer where latent systems could catch up. And that’s where the cracks began to show—because for all the progress the Department has made in seeing the enemy, it still struggles to see itself in real time.

Take air-defense interceptors, the multi-million-dollar missiles critical for protecting U.S. forces, facilities, and allies. Data on where interceptors are, what condition they are in, and whether they have been expended is spread across multiple legacy business systems that are not integrated. At any given moment, the Department does not know how many of these critical munitions it has or where they are. Shortly after Epic Fury began, the Department found it could no longer rely on its business systems. It reverted to phone calls, spreadsheets, and paper-based reporting.

Over the past decade, the Department has integrated the kill chain to make the soldiers working in targeting cells one hundred times more effective. During that same time, business systems that support critical functions across the Department have barely evolved. In many ways, they’ve become less integrated, intermediated by data lakes and layers of analytic tools that separate them from the Department’s warfighting mission.

Within a warfighting environment, the Department’s business systems do not provide relevant data that they were designed to track, including: how may aircraft are operational, where units and service members are, and if it has the munitions needed to supports its plans. Operation Epic Fury showed us that the systems that support critical functions—managing personnel, equipment, and munitions—cannot live on a back-office island; they are a subset of warfighting systems that are inextricably linked to the Department’s core mission.

Know the Enemy and Know Yourself

We often talk about the fog of war as something imposed by the enemy—something external that obscures clarity and complicates decision-making. Epic Fury suggests something different: that a meaningful portion of that fog is self-inflicted.

Sun Tzu’s maxim to “know the enemy and know yourself” is oft-repeated but rarely interrogated. We have spent decades investing in the first half of that equation. Our ability to sense, track, and understand adversary behavior has improved dramatically. But our ability to understand our own posture—our true readiness, our real constraints, our capacity to act—lags behind.

You can see this most clearly when you imagine the moment of decision. A commander is planning a strike package. The target data is current. The threat environment is continuously updated. Weather inputs are live. That entire side of the equation has been engineered for speed and precision. But when attention turns inward, the picture degrades. Munitions data trails reality. Fuel status reflects yesterday’s conditions. Maintenance readiness is fragmented across systems. Everything about the enemy is immediate; everything about ourselves is historical.

The discrepancy between how we fight and how we understand our ability to fight is a direct result of the fragmented way we designed our systems. Conway’s Law has shaped the Department more than any formal architecture. Systems mirror the structures that build them, so we ended up with Army systems, Navy systems, and Air Force systems, as well as logistics systems, financial systems, and personnel systems. Each system is aligned to its organizational owner, optimized for its specific function, and evolves on its own timeline. That approach made sense when coordination happened through meetings and staff processes and forces were sustained by nearby stockpiles of materiel. It does not hold in a world where decisions are distributed, timelines are compressed, supply chains stretch across continents, and outcomes depend on the seamless interaction of all parts of the enterprise.

What Epic Fury underscores is that the need for speed is no longer confined to warfighting systems. The Department largely has the data it needs, but it lacks the ability to act on it. Many of the systems that underpin readiness continue to operate on timelines measured in days or weeks. Systems that require manual intervention at every step are fundamentally misaligned with the speed of modern operations. Attempting to compensate with dashboards and reporting layers has not resolved this mismatch. If anything, it has made it more tolerable by providing an acceptable level of visibility during peace time—allowing programs to claim progress that’s “good enough.” Storing and displaying information is too low of a bar. Critical systems must participate in operations, translating data into warfighting decisions.

This leads to an obvious conclusion: if a system does not function in a warfighting context, it does not function for the Department of War. Compliance, reporting, and even efficiency are poor proxies for the true objective: fight and win. Systems that cannot support that objective under realistic conditions are liabilities.

Fight and Win

The separation between business systems and warfighting systems is an artifact of the past and now a critical risk in the present. This is a problem that requires a solution more fundamental than “modernization” in the traditional sense. Instead it requires a shift in how these systems are conceived, built, and evaluated.

They must be tied directly to mission outcomes, fully integrated into warfighting platforms, and exercised in conditions that resemble real operations. They must be designed to operate at the speed of the fight, not the speed of administrative process. Like warfighting systems, they must speed the integration of AI to accelerate and improve decision making. They must provide a continuous, accurate understanding of our own posture—because without that, every other advantage begins to erode.

There is no longer a meaningful distinction between the systems that run the business and the systems that fight the war. There is only the war machine itself, and in the end, it either enables the mission—or it does not.

KEN BIBERAJ is the host of Coffee with Ken and advises advanced manufacturing and defense companies on site selection and industrial strategy.

A few weeks ago, I sat across from Madeline Hart—co-author of Mobilize: How to Reboot the American Industrial Base and Stop World War III—at the inaugural Coffee with Ken Summit just outside D.C. There were 300 people in the room: senior officials from DoW, DOT, and DOE; aerospace and defense industry leaders; builders, investors, and operators who are all trying to figure out the same thing. How do we actually do this? How do we take the national will to reindustrialize and turn it into factories on the ground?

It’s clear that the will to mobilize is there. The capital is there. The companies are ready to build. But they keep running into the same wall.

From my perspective as a real estate site selector, that wall isn’t in Washington. It’s in the fifty states that control every piece of land, every power line, every permit, and every zoning decision that determines whether a factory gets built. Mobilize is a book about the national will to build. This is an essay about who is responsible for the execution once that will exists.

For Once, Don’t Blame the Swamp

Washington cannot rezone a county. It cannot accelerate a permitting process administered by a state environmental agency. It cannot extend a water line, upgrade a substation, or entitle 3,000 acres for an advanced manufacturing campus. It cannot override local opposition to an industrial facility or build the workforce pipeline that makes a facility viable for the next thirty years.

All of that—every single element that determines whether a site is shovel-ready and ribbon-worthy—is controlled at the state and local level. I advise advanced manufacturing and defense companies on where to locate, and I watch this dynamic play out in real time across the country.

The gap between federal ambition and state execution is a billable problem that I make a living working to navigate. When a site timeline slips by two years because a state agency couldn’t coordinate a utility extension with a permitting office, it is a capacity failure three levels down, in a building most people in DC have never heard of.

The states that understand this are pulling away. The states that don’t are losing deals they don’t even know they’re losing.

The Incentive Playbook is Outdated

For forty years, economic development has run on a simple formula: recruit a company, offer a tax abatement, cut a ribbon, issue a press release. Cash grants. Job creation credits. Maybe some workforce training dollars thrown in. It was as simple as plugging some variables into an Excel model.

That model worked well enough for a world of conventional manufacturing and distribution facilities. It does not work for the AI-powered factories of the future.

A gigafactory doesn’t need a tax credit. It needs a fully entitled site with rail access, 300-plus megawatts of dedicated power, a water supply commitment, and a permitting timeline that won’t slip three years into the project because of a challenged environmental review. An advanced semiconductor fab needs a workforce pipeline that doesn’t exist yet, a supply chain ecosystem in the same region, and a state government sophisticated enough to coordinate across a dozen agencies without dropping the ball.

Most state economic development agencies are not built to deliver any of that. They are built to process incentive applications for Fortune 200 companies and host trade missions. Those are not the same thing, and companies building the next generation of American manufacturing know the difference immediately. This new group of innovators is venture backed, moves fast, and did not exist three to five years ago. States have a hard time understanding these companies, much less underwriting them.

The Data Center Tradeoff

There is a specific distortion worth naming: the data center.

Over the last decade, states quietly became addicted to data center investment. The deals were easy and involved minimal permitting complexity relative to capital intensity. In turn, these capital-intensive projects looked great in press releases. The problem is they consumed enormous amounts of power and land, and generated almost no employment density per acre or per megawatt.

Now, as manufacturing companies come looking for megawatts and sites, they are finding grid capacity already committed, the best industrial land absorbed, and incentive budgets exhausted on facilities that will never produce a meaningful number of jobs. It is hard to blame states entirely—they didn’t have a national industrial policy framework to tell them what they were trading away. But the tradeoff is real, and it is biting.

I raised this at a conference not long ago. I mentioned that the president’s national security strategy explicitly calls for reindustrialization, and I asked the room full of state and local economic development officers what they were doing to meet that national security need. The blank stares didn’t go unnoticed. Dozens came up afterward asking what they were supposed to be doing.

To be fair, the data center debate is not black and white. AI-powered factories will require data centers, even while they compete with traditional industrial investment for skilled labor, land, and resources.

These complex decisions will be made in the zoning boards and utility commissions and permitting offices of fifty states that were never told they are the execution layer of a national mobilization. We are trying to run a World War II-scale industrial build-up through a system that doesn’t know it’s at war.

What Good Looks Like

I travel this country constantly—not to attend conferences, but to sit down, have real conversations, and learn what’s actually happening on the ground floor of reindustrialization. What I find is genuinely encouraging. There are communities doing this right, but they aren’t getting nearly enough attention.

Cleveland is one of them. The city-backed Site Readiness Fund has done something most cities talk about but never do: acquire hundreds of acres of real estate. Cleveland is now doing the hard, patient work of taking those sites to market—determining what industries they are positioned to support, reducing upfront costs for companies, and building the infrastructure that makes a site real rather than just a promise on a map. That is exactly the kind of institutional investment that changes a community’s competitive position by putting it on a path to have factories up and running inside of three years.

Huntsville might be my favorite example. Yes, they have more PhDs per capita than almost anywhere in America. Yes, they landed Space Force. Yes, the innovation ecosystem there is genuinely world-class. But what actually makes Huntsville work is simpler than any of that: an accessible, hands-on mayor; housing being built at a pace that matches demand; and a community culture that feels alive. I went back with two of my sons for Space Camp a few years ago. When a city is good enough that you bring your kids back on vacation, they’re doing something right.

Then there’s Tulsa, which honestly shocked me. I didn’t know what to expect. What I found was a community that has figured out how to get the public sector, private sector, and philanthropic sector to work together. A lot of that connective tissue runs through the George Kaiser Family Foundation, which has woven together the kind of cross-sector partnerships that speed up market access and reduce uncertainty for companies.

And Pittsburgh. I dragged my family there for summer vacation last year. We put the kids in robotics camp at Carnegie Mellon and visited more museums than we did on a trip to Paris. But the story that really gets me is the city’s Hazelwood Green project. Tishman Speyer is transforming the former Jones and Laughlin Steel site—one of the most heavily guarded industrial facilities in America during World War II—into a mixed-use innovation district anchored by Carnegie Mellon’s robotics program and the Pittsburgh Steelers. The most important steel site in the American war effort is becoming the address for American robotics. That’s a blueprint for what reindustrialization can look like when a community puts manufacturing at the forefront.

What States Need to Do

The path forward is not complicated in theory:

Invest in sites before companies show up. The single greatest competitive advantage a state can offer is a development-ready industrial site: entitled, powered, watered, and connected. This requires capital investment without a guaranteed return and a multi-year time horizon that is politically uncomfortable. Cleveland is doing it. A handful of others are doing it. They are winning deals that the rest of the field doesn’t even know they’re losing.

Reduce the upfront cost and risk of locating a facility. Companies deciding where to build right now are choosing based on execution certainty more than incentive packages. A state that can credibly commit to a twelve-month permitting timeline, a shovel-ready site, and a coordinated infrastructure delivery is worth more than a state with a bigger tax credit and an uncertain timeline.

Stop competing on incentive volume and start competing on ecosystem fit. The most durable manufacturing investments are embedded. A company that builds where it fits the regional talent pool, the supply chain geography, and the workforce culture will be there in thirty years. A company that came for the abatement will leave when a competitor state offers a better deal. Tulsa gets this. Huntsville gets this. Most states are still chasing every RFP that crosses their desk regardless of fit.

Build real institutional capacity to execute. This means dedicated project management infrastructure inside government—a team that manages complex projects across agencies and drives them to completion instead of merely processing applications. Pre-negotiated utility agreements. Pre-cleared environmental frameworks for defined industrial zones. A single point of accountability. The difference between a state that can execute and one that can’t is almost never about intent. It’s about the organizational infrastructure to follow through.

And stop giving all the land and power to data centers without an industrial strategy that explains why.

The National Stakes

The premise behind every dollar of federal money committed to industrial policy is that the United States must rebuild its capacity to produce what national security and economic resilience require, from semiconductors to batteries.

This battle will be won or lost in unexpected places like county commission meetings, state utility commission hearings, and permitting offices currently staffed for a quieter era. The adversaries we are competing with do not face those constraints. They build what they decide to build, where they decide to build it, on the timeline they set. Meanwhile, we route major industrial investments through a multi-year permitting gauntlet with unpredictable outcomes and call it a free market.

The federal government has begun working to open federal land for data centers to meet AI demand, which is a start, but a similar strategy should be put in place for broader advanced manufacturing. As a site selector, I push my contacts in DC constantly for better access to federal sites, and the bureaucratic friction is real. If reindustrialization is really a national security concern, our site selection should not be limited to what states offer and private owners are willing to sell. More of the federal estate should be accessible for this mobilization effort.

We need to be able to build fast, pivot fast, and upgrade fast. Right now, our states—the organizations that control the land where those factories go—are not structured for that tempo. We need a real industrial policy. Not a collection of federal funding programs, but a genuine alignment of national goals with the state-level execution mechanisms that determine whether those goals are achievable. That means treating state economic development capacity as a national security asset and investing in it accordingly. It means governors willing to restructure agencies, allocate capital to site preparation, and make long-term bets that don’t pay off until the next administration.

One More Thing

I’m optimistic about America because of the time I spend on the road in places like Cleveland, watching a city invest in its own industrial future. And Huntsville, where a mayor who picks up the phone has helped build one of the most dynamic innovation ecosystems in the country. And Tulsa, where I watched public officials, foundation leaders, and private investors finish each other’s sentences because they had been in the same room long enough to understand each other. And Pittsburgh, watching my kids learn to code robots in a building two miles from where American steel won a world war.

These places are proof that rebuilding is possible. The question is whether the rest of the country—and the state and local institutions that hold the keys to reindustrialization—will move fast enough to join them.

Who will build the arsenal of the future? Storied, legacy defense manufacturers? Buzzy, VC-backed startups? Firms and founders yet unknown?

This is a hot topic in Washington and other hubs of power. Many careers and many billions of dollars hinge on the answers—not to mention many lives.

And so the question provokes passionate reactions. Armin Papperger, chief executive of the German defense firm Rheinmetall, issued a salvo on behalf of the legacy defense firms earlier this year. He criticized small defense manufacturers in Ukraine, which are producing waves of lightning-fast drones to patrol the front line and intercept incoming attacks. These manufacturers are "play[ing] with Legos," he told The Atlantic’s Simon Shuster. Their factories were staffed by "Ukrainian housewives" with "3-D printers in the kitchen." "This is not innovation," he concluded with Teutonic bluntness.

The response was swift and merciless. Volodymyr Zelenskyy quipped that "if every Ukrainian housewife can really produce drones, then every Ukrainian housewife could also be the CEO of Rheinmetall." Wags on X designed patches for a fictitious "17th Housewife Drone Regiment." Rheinmetall quietly issued a clarification praising "the innovative strength and the fighting spirit of the Ukrainian people."

Papperger’s outburst is one salvo in the ongoing cold war between legacy defense manufacturers and new competitors. The implication of his comments is that only established companies (like his) with meticulously tested and validated products can deliver the capabilities that modern war demands.

This claim shouldn’t be dismissed outright. After all, the war in Ukraine is not merely a war of drones and AI; it is also a war of armor and artillery, the very things built by companies like Rheinmetall. Other recent fights have been even more lopsided in favor of legacy companies and their products. It wasn’t startups that supplied America’s military actions in Venezuela and Iran. It was Black Hawk helicopters that ferried the Delta Force to Nicolas Maduro’s compound, F-35s and Tomahawk cruise missiles that battered the IRGC, and Patriot and THAAD batteries that protected our troops and allies from Iranian missiles. Novel capabilities made appearances, but they were supporting players rather than leads. Western militaries still rely heavily on the exquisite systems provided solely by legacy defense firms. They will for years to come.

Behind Papperger’s bravado, however, one senses the unease of a buggy whip maker who has seen a horseless carriage chugging around the bend.

Legacy defense companies are more vulnerable perhaps than they have ever been because they have no good answer to the central dilemma that we in the West face: our weapons are expensive and scarce, while our enemies’ weapons are cheap and plentiful. In all but the briefest and most lopsided campaigns, our side burns through its stockpiles of high-end munitions and interceptors faster than it can replace them. Our enemies are counting on it.

We have known this for years, but the roughly month-long campaign against Iran made the problem impossible to wish away. The United States and its allies expended enormous quantities of precision interceptors and stand-off missiles—systems that cost hundreds of thousands to millions of dollars apiece, built by legacy defense companies. They worked. (For the price, they should.) But there simply are not enough of them.

The Center for Strategic and International Studies estimates that the U.S. military expended roughly one-third of its Tomahawk cruise missiles, one-fourth of its JASSMs, and more than one-half of its Patriot and THAAD ballistic-missile interceptors. The lead time to replace those munitions is measured in years, creating a conspicuous window of opportunity for our adversaries.

This dawning reality has set off a mad scramble to rearm—and to find alternatives to existing systems—which is both opportunity and threat for legacy defense firms. The president’s $1.5 trillion defense budget includes tens of billions of dollars for new exquisite missiles and missile interceptors. But it also invests heavily in cheaper alternatives, potentially circumventing the legacy defense complex altogether. The buzz word of the moment is “attritible mass,” a phrase that translates roughly to “lots of weapons, at a lot lower price.”

So suddenly, legacy defense firms face the prospect of serious competition against dozens of “housewives” for defense funding and contracts. And they are starting to feel the heat.

The Secretary of War released a video that vividly illustrates the new reality of defense contracting. He stated that the Pentagon is prepared to offer steady, long-term contracts to refill its arsenal, but it expects legacy defense manufacturers to invest in new factories using their own money, to boost output of key munitions, and to hold prices steady. “If they fail to deliver,” the secretary warned, “we will hold them accountable and bring in new companies who will.” The video depicts legacy manufacturers as paunchy older men in pinstriped suits and the new entrants as jocular tech bros in vests.

Acquisition dollars are beginning to shift to new entrants. The Department of War recently announced agreements with four defense companies, including Anduril, to supply 10,000 low-cost, containerized cruise missiles over the next three years. It reached a similar agreement with Castelion to supply at least 1,000 hypersonic missiles. The Pentagon’s press release boasted that every company involved is a new entrant. No primes allowed.

And although it is still early days, new entrants’ weapons are already proving their mettle on the battlefield and the test range.

Ukraine’s war of national survival has proven an excellent incubator and proving ground for new defense firms. A few short years ago, Papperger wasn’t entirely wrong: Ukraine did indeed rely on homemade drones made in the military equivalent of cottage industry, to the point that it launched a “People’s FPV” initiative to train non-technical workers in the relevant skills. Now the industry has matured through a combination of bottom-up innovation and top-down support from programs like BRAVE1, a government-backed effort to nurture and coordinate the country’s startups. According to the Snake Island Institute, Ukraine’s defense tech ecosystem grew since Russia’s invasion from roughly a dozen companies to more than 1,500. These companies have enabled Ukraine to scale its domestic drone production from roughly 3,000 annually at the war’s beginning to an astounding 5 million per year today.

Now the most successful of these firms are raising private capital, hungrily eyeing the broader global defense market—and dominating head-to-head trials. The housewives have learned a thing or two about what works and what doesn’t from four years of existential warfare. Increasingly they are winning on quantity and quality, as the proximity of producers and front-line operators creates a rapid feedback loop for improvement.

The products of these new entrants are already popping up in Western military service. In recent Congressional testimony, Army Secretary Dan Driscoll lauded the Merops—a small, inexpensive interceptor drone developed by an American company and battle-tested in Ukraine—as a breakthrough solution to shooting down Iranian drones in the Middle East. Scarcely a week into Operation Epic Fury, the Army purchased 13,000 Merops at roughly $15,000 apiece. At scale, the price is expected to drop below $10,000. The drones are being used to knock down Shaheds that cost $30,000 to $50,000 each, putting the United States—for once—on the right side of the cost curve.

All of this is real competition, and it is forcing legacy firms to adapt in several ways. Since the start of Epic Fury, prime contractors have reportedly agreed to triple or even quadruple production of key munitions and interceptors. They are racing to modernize antiquated factories and manufacturing processes, bringing 21st-century technology to lines stuck in the past.

They are also trying to develop their own answer to attritible mass, from small drones to affordable cruise missiles. One especially ironic example from Europe: late last year, Rheinmetall expanded its offerings of small kamikaze drone. "Lego"-like drones may not be Papperger's idea of innovation, but it seems they will play a role in the future even of his company.

So who will win this high-stakes, high-dollar contest for the future of the Western arsenal? The answer may not be a clean either/or. Western militaries clearly need a mix of high-end systems for range and precision and low-end systems for endurance. But in each segment of the market, the firms that win will have to earn it through speed, the ability to iterate, and the ability to ship product in the quantities and at the price point that Western governments demand.

The firms that succeed in this new paradigm will be the ones that channel their inner housewife.

DOMINIC VENTIMIGLIA and JARRETT LANE were teammates in the Middle East over ten years ago. Dom retired from the U.S. Army Special Forces as a Chief Warrant Officer 5 and is now a Principal at Squadra Ventures. Jarrett is Chair and Co-Founder of the N.C. Critical Technologies Alliance.

At the height of the war with Iran, the regime systematically struck U.S. military radars, communications infrastructure, air-defense systems, and other assets across the Middle East with low-cost drones — basically cheap missiles and flying improvised explosive devices. Seemingly surprised by the Iranians’ response, the U.S. military moved THAAD systems and Patriot batteries from the Indo-Pacific into the Middle East to guard against these relentless drone attacks. The United States and its partners in the region quickly depleted their supplies of expensive, hard-to-produce munitions to shoot down cheap Iranian drones.

A common refrain at the moment is that the U.S. military failed to learn lessons from the Russia-Ukraine war, as if drones did not emerge as a real threat until the last four years. It is hard to excuse the Department’s slow reaction to the innovations in drone warfare the world has witnessed since the Russia-Ukraine war started in 2022. However, focusing on the failures to learn from Ukraine actually gives the Department too much benefit of the doubt.

In reality, commercial, improvised, and low-cost kamikaze drones were a known threat long before Russia and Ukraine began using them. Iran has prioritized its drone program as a core component of its asymmetric military strategy since at least the 1980s. The Shahed has been part of Iran’s drone fleet since 2012, and Iran has used kamikaze drones since at least 2015. ISIS weaponized commercial drones across the Middle East, including against U.S. forces.

While assigned to the U.S. Central Command between 2015 and 2017, we saw first-hand the threats drones posed to U.S. servicemembers and critical assets across the Middle East. Our mission was to simulate attacks by adversarial special operations elements to test the security and resilience of U.S. military installations and critical infrastructure. We applied known and emergent adversarial tactics and imagined potential new tactics to identify courses of action that an enemy might realistically pursue. Drones quickly became one of our go-to tools.

Unfortunately, many were slow to appreciate the threat this new tool posed. In our experience, the idea of using cheap drones to degrade or dismantle critical capabilities was typically met with one of three responses. The first was incredulity. To some, drone attacks posed a nominal force protection problem because they were crude and usually unsuccessful. There was little appreciation that new forms of warfare usually begin crudely. The second response was dismissal. Drones could be easily dealt with by shooting them down or taking out the operators. The third response was acknowledgement that drones were a problem—but that they were somebody else’s problem to solve.

To be clear, our experience was not unique. We are far from the only people who anticipated the drone threat and called attention to it. At the time, the Joint Improvised-Threat Defeat Organization, or “JIDO,” which operated under the Defense Threat Reduction Agency, was working with “a sense of urgency” to field counter-UAS capabilities.

And yet, here we are: Iran successfully targeted the backbone of the U.S. military’s operational capabilities across the Middle East. As Iran moved up its escalation ladder, commercially-owned infrastructure that the U.S. military may use in the region was also targeted. Though more Iranian drones have been intercepted than not, as one outlet reported, the number of successful drone strikes on U.S. military assets and infrastructure reveals “a pattern of inadequate protection for strategic locations.” The U.S. military’s unpreparedness is particularly troubling when one considers that Iran’s response represents just a fraction of what China could employ to disrupt and degrade mission-essential capabilities.

Such unpreparedness is not just an operational failure; it is the symptom of systemic breakdowns in policy, process, and programs that should be helping the Department get ahead of threats and field appropriate solutions. While fielding counter-UAS solutions is critical, addressing the systemic challenges must be done, too, for the Department to not get caught flat-footed by new threats again. Drawing from our own experience, we have a few recommendations to offer and thoughts on what else is to come.

Feedback loops from rigorous exercises and red teaming can help the Department foresee threats and proactively develop solutions.

The wars in Ukraine and Iran should create a sense of urgency within the Department to build, scale, and institutionalize mechanisms to identify new threats from the tactical edge, rapidly iterate solutions, and push them into the field. These mechanisms should extend beyond just capturing lessons from troops in contact. Exercises that meaningfully stress-test operational proficiency, resilience, and adaptability — as opposed to running exercises focused on relationships and posturing — can help expose weaknesses and threats. Those exercises must also be integrated with acquisition processes. An after-action report isn’t good enough — senior leaders almost never read them. Acquisition executives and technical experts should take a front-row seat as observers in these exercises to see for themselves the problems that need solving. First-hand understanding can help leaders drive top-down decisions through the Pentagon’s bureaucracy about how to get ahead of emerging requirements and threats.

The Department should also revitalize and expand red-teaming functions to expose vulnerabilities in multiple domains. One hub of such activity, the Army’s Asymmetric Warfare Group (AWG), was shuttered in 2021 despite its successes. The AWG provided advice and support to Army commanders at the tactical and operational levels. It was celebrated for its work developing solutions to threats like IEDs. The AWG was geared more for the age of counterterrorism than great-power competition. Yet the decision to close the AWG instead of updating its mission implies an assumption that countries like China and Russia would not wage asymmetric warfare against the United States and our allies and partners. This assumption could not be more wrong.

The wars in Ukraine and Iran show that dismantling the AWG was a mistake. Further, asymmetric warfare isn’t just kinetic: China and Russia are doubling down on other methods, including cognitive warfare, precisely because of its asymmetric advantages. To the authors’ knowledge, nothing has functionally replaced the AWG. The Department should reconstitute the AWG, or some variation of it, with a remit that reflects the fact that asymmetric warfare is not just the tool of violent extremists; it’s an increasingly sophisticated, multi-dimensional discipline used daily by our most capable adversaries.

Further, the multi-domain nature of asymmetric threats from our adversaries requires a joint perspective, not just solutions through the lens of one Service. The Department should also think anew about how to use red teaming to map and understand interdependencies of critical assets and networks — military and commercial alike — so it can make informed investments into protection measures and build the redundancy needed for the modern threat environment.

The qualities and experience of the personnel that comprise a joint AWG 2.0, or similar red-teaming functions, are essential to the organizations’ efficacy. The teams should be capable of thinking creatively and should be empowered to do so. They should draw on skill sets from special operations, the intelligence community, social engineering (e.g., social media expertise), and engineering and product development (e.g., rapid prototyping).

Finally, as with exercises, robust feedback loops are foundational to the red teams’ ultimate success. For the Department to gain real value out of red-teaming activities, feedback loops must inform institutional learning and strategic decision-making. Learnings should not just inform protection measures or tactics at the unit, installation, or theater level. Rather, learnings need to translate into new resources, updated theater and global-level contingency plans, and even training and doctrinal revisions.

One potential approach to bridge the gap between red teams’ operational and tactical learnings and strategic decisionmaking is to directly involve elements of the Office of the Secretary of Defense (e.g., Strategy, Plans, and Forces) or the Joint Staff. For example, OSD or the Joint Staff could provide direction on what vulnerabilities they want tested. They could also be the primary consumers of red teaming results.

Translating learning from partners and technical experts into action will be critical to stay ahead of future threats.

Despite Ukrainians’ clear expertise in drone production, drone operations, and counter-drone capabilities, the U.S. government was slow to accept help from Ukraine, even as Iranian drones began striking assets across the Middle East. The United States ultimately made the right call by welcoming Ukrainian drone experts to help protect military bases in the Middle East. These sorts of cooperative efforts and knowledge exchanges should be encouraged going forward. Learning from allies and partners — especially those with first-hand experience of new threats, technologies, and tactics — will be crucial for the U.S. military to stay ahead of emergent challenges.

Ongoing changes to the Department’s security cooperation enterprise present an opportunity to improve our own learning from allies and partners. Aligning the Defense Security Cooperation Agency—the organization responsible for arms transfers—to the Undersecretary of Defense for Acquisition and Sustainment could create a clearer through-line for learnings from allies and partners to the Department’s acquisition leadership.

The Department should also send teams to interface with allies and partners explicitly for the purpose of gleaning and exchanging lessons learned. To date, very few personnel have been assigned to systematically learn from Ukraine. U.S. military personnel assigned to embassies have an important role to play in helping facilitate learning from allies and partners, but these personnel are already task saturated promoting weapons sales, facilitating combined training, and more.

Drones abroad are just the start. The United States is vulnerable at home and new threats are rapidly emerging.

Drone threats will continue to spread and intensify. Without serious attention to domestic counter-UAS policies and solutions, the United States remains vulnerable to an Operation Spiderweb-esque event here at home. A recent incident in which waves of jam-resistant drones reportedly flew over Barksdale Air Force base show just how far the government still has to go in developing a coherent counter-drone strategy and response options.

The threat is not just purpose-built military drones. Commercial off-the-shelf drones and drones custom-built with commercial components pose a serious danger, too. For example, Houthis are using commercial micro-jet engines to build fast drones with significant range. And drones are just one example of how commercially available technologies pose a serious threat.

While the Department is still adjusting to using commercial off-the-shelf solutions, adversaries, violent extremist organizations, and criminal groups are using them with greater and greater effect. Electronic warfare technologies are increasingly available on the open market. Criminal organizations are already using signal scramblers and GPS jammers to disrupt law-enforcement communications. For a few hundred dollars, just about anybody can buy the necessary parts to build a crude anti-aircraft missile or shoulder-fired rocket. Autonomous and semi-autonomous systems, including robotics and maritime systems, will increasingly threaten U.S. military personnel and assets, as well as commercial infrastructure like ports, power generation, and oil and gas. Biological threats, including weapons concocted in illegal labs here in the United States, could be used to infect human populations or agriculture.

Artificial intelligence will amplify the threat by making the production and coordination of these technologies easier. For example, consider human-machine integration. Imagine human operators seamlessly interfacing with swarms of drones and robotic systems — commanded with nothing more than thought — using widely-available technologies. It sounds like science fiction, but it is coming. Three years ago, the IEEE published an article covering how the costs of “open-source neuroscience hardware” are coming down. Expect that trend to continue.

Sticking with human-machine integration as an example, the Department should be collaborating with technical experts in startups and universities to red-team scrappy human-machine systems built using only commercial off-the-shelf and open-source technologies. Operational experts in the special operations community, for instance, can apply their know-how to imagine new types of complex attacks or battlefield awareness systems.

The Department should learn from its drone failures to avoid future surprises.

The ultimate lesson is not simply about drones. It is about a system that struggles to recognize, prioritize, and act on emerging threats before they scale. Fixing that system, not just fielding the next countermeasure, is what will determine whether the United States can compete in an era defined by rapid, asymmetric innovation. Re-investing in teams and programs to forecast future threats and, most importantly, listening to and acting on the expertise of others, can go a long way toward making sure the Department isn’t caught flat-footed by the next threat.

The next disruption is already taking shape; it will not wait for us to catch up.

We often default to a particular version of World War II. It’s the one everyone knows: D-Day, carrier battles in the Pacific, armored divisions pushing across Europe. It’s a story about tactics and decisive moments. The Band of Brothers war, you could call it.

But a lot of the war was decided somewhere else entirely. In places like Detroit and Pittsburgh, production lines ran around the clock and output became the limiting factor of what was possible on the battlefield. It was factories that determined the pace and scale of the fight.

When Franklin D. Roosevelt called the United States the “Arsenal of Democracy,” he was describing how the system actually worked. Combat power started in the factory and manifested on the battlefield. We understood that then. We don’t operate that way now.

Today, the conversation is centered on command and control decision advantage—how quickly we can move from sensing to decision to action. Systems like CJADC2 are built around that idea, emphasizing faster decisions, tighter OODA loops, and more connected forces. That matters, but in today’s word it’s not the constraint most people think it is. Just look at CENTCOM, which has executed over a thousand targets a day in Operation Epic Fury.

The constraint is production, and you can see it in how we fight today. It’s entirely possible to have a good operational picture and still not have a clear answer to basic questions like: how many munitions do we have ready right now? How many munitions can we produce? The data exists, but it’s spread across systems, updated on different timelines, and owned by different organizations. Instead of seeing the answer, we assemble it, which slows everything down in ways that only become clear in the fog of war.

One part of the system moves at machine speed, while another part moves at human speed. We’ve built a very good kill chain, but we haven’t built the equivalent system for production and sustainment. Once the opening phase passes, the question changes from who can strike first to who can keep going—who can replace what they use, shift production, and absorb disruption.

Enter WAR MACHINE

The solution is to build a system that enables real visibility and control of defense industrial supply chains and production. We call this WAR MACHINE. It’s not a program or a piece of software, but a way of looking at the system as a whole. The factory, the suppliers, the depots, the logistics network, and the units using the equipment are all part of the same system. When one part slows down or breaks, the machine doesn’t work as it should.

Right now, those parts don’t operate as one system. The primes have their own view, sub-tier suppliers have another, and the government has a third. Each one sees a slice, but no one sees the whole picture in real time, which makes it difficult to answer straightforward questions about constraints, bottlenecks, and spikes in demand.

It also changes the dynamic between the Department and industry. Rather than wield the industrial base, the Department is at its mercy. The DoW often has less visibility into a given program’s supply chain than industry. That’s an unacceptable status quo during peacetime, and it becomes existential during a conflict. It’s ironic that the DoW requires defense contractors to open up their books and comply with Cost Accounting Standards so auditors can pore over pennies, yet it does not apply the same scrutiny to the area that would really make a difference: production. Visibility into a weapon system’s production should be a requirement, just like payload capacity, range, or accuracy.

You hear the pushback on this proposal almost immediately: the Department shouldn’t have this level of visibility into industry, lest it overreach and cross a line into controlling how companies operate. Such an argument ignores how competitive hardware products are manufactured in the commercial economy. Apple does not take a laissez faire approach towards its suppliers. Quite the opposite. Apple rear deploys its engineers to all of its suppliers. It’s a similar story with the automotive OEMs and their relationship to the Tier II and Tier III suppliers. It’s helpful to think of Apple and the auto companies as the first customers of their products. Demanding factory floor visibility from their suppliers is a matter of customer service. It follows that when the stakes are higher than luxury goods, the DoW should not operate in the dark for life-and-death production runs of artillery.

The reality is that the defense industry’s truculence on this issue is a bit of “the lady doth protest too much.” The defense companies themselves often have poor visibility into their own supply chains and are none too eager to expose such a state to their customer. In turn, suppliers to the defense companies are often reluctant to disclose bad news like parts delays. All the more reason for the DoW to demand more of industry as a forcing function to make the entire industrial base more effective.

There’s a carrot in there for industry as well. Today, companies are often planning production against incomplete or lagging signals from the government—uncertain demand, shifting priorities, and funding volatility all get in the way of making clean decisions. We ask industry to surge, but we don’t give it a clear picture of what “surge” actually means over time. A system that connects the Department to its industrial base would not only give government visibility into supply, it would also give industry visibility into demand, allowing both sides to operate off the same reality instead of two different approximations.

Recent operations make the visibility gap harder to ignore. In Operation Epic Fury, U.S. and partner forces executed one of the most intense strike campaigns in decades, consuming large quantities of precision munitions across multiple systems. The operational side performed exactly as designed—fast, coordinated, and effective. But significant portions of key munitions inventories were depleted, and rebuilding those stockpiles will take years, in some cases.

Innovation in bits advanced faster than atoms, which means awareness and targeting far outpace production capacity. The limiting factor is whether the industrial base can regenerate combat power at the pace the fight demands. The next conflict—especially if it’s in the Indo-Pacific—is unlikely to be a brief exchange. It will be defined by distance, scale, and sustained demand across air, maritime, and missile systems. It will stress supply chains in ways we haven’t experienced ever. It will expose every fragile node, every single-source dependency, and every place where production can’t keep up.

Industrial Power is Combat Power

In World War II, the solution was alignment. The entire system—government, industry, and workforce—moved in the same direction. Arthur Herman’s Freedom's Forge captures how leaders like William S. Knudsen and Henry J. Kaiser coordinated production across companies and supply chains. It wasn’t elegant, but it worked because the factory and the front line were part of the same system.

We’ve seen a version of this play out in the commercial world, as well. In the book American Icon, turnaround CEO Alan Mulally walked into Ford Motor Company when it was close to failure. One of the first things he did was force visibility across the enterprise and into the supply base. Ford soon understood what was happening inside its operations and across its suppliers, issues surfaced early, and leaders could act before problems cascaded. That shift—seeing the system clearly and acting on it—was a big part of what kept Ford from going bankrupt, even while General Motors and Chrysler went under.

This is the shift the Department needs to make. Creating a system that sees itself clearly enough to act transcends building more reports or better dashboards. We must connect production, supply, and sustainment the same way we’ve connected sensors and shooters, so that constraints are visible early and decisions about tradeoffs can be made with a full picture instead of partial views stitched together over time.

The underlying point is straightforward: industrial power is combat power. In a conflict that lasts more than a few weeks, the side that wins is the one that can keep producing, keep adapting, and keep supplying its forces under pressure. The path from factory to foxhole is as much a part of the fight as any kinetic action.

Right now, many still assume the industrial base will keep up, supply chains will hold, and production will surge when the time comes. Operation Epic Fury proved those assumptions false. WAR MACHINE is the system we need to build now if we expect to fight, sustain, and win any future conflict.

Peter Mitchell is a U.S. Army officer and former instructor of strategy at the U.S. Military Academy at West Point. His novel Captain Hawke will be released this summer. He has also written an adventure following a young Clausewitz, a story that, much like Scharnhorst’s early efforts, is still searching for its breakthrough. Follow him on X @peternmitchell.

In 1801, a Hanover-born artillery officer named Gerhard Scharnhorst arrived in Berlin with an unusual mission from the King of Prussia: he was tasked with making the Prussian Army think.

This was, at the time, a revolutionary act.

Quite literally—people thought he was a crypto-republican.

Prussia in 1801 was (to her officers’ minds) already perfect. Her army was the most storied in Europe. It stood upon the legacy of Frederick the Great, who had died in 1786 having won wars that defied all considerations of manpower and geography. Prussian officers still dressed after his fashion, drilled in his formations, and repeated his maxims as gospel. The army had become a living museum, and in such a place as that the most dangerous thing an officer can do is suggest that the exhibits be removed.

Gerhard Scharnhorst was just such a man.

The Parade Army

To understand what Scharnhorst was up against, you have to understand how badly the Prussian Army had stagnated by the dawn of the nineteenth century. Structurally, it was still the army of Rossbach and Leuthen of five decades before: fighting in rigid linear formations with brutal drill discipline enforced by the rod. Officers were selected by birth rather than ability and operated in a command culture that punished initiative as a form of insubordination. The enlisted ranks were filled with levied peasants, criminals, vagabonds, and foreign mercenaries held together through coercion rather than cohesion. The whole edifice assumed that war would obligingly be fought by the same immutable Frederician script.

However, Napoleon Bonaparte had written a new one.

South of the Alps, France had spent the better part of a decade kicking the armies of the old order up and down the Po Valley. The levée en masse had mobilized the whole of French society and replaced small, professional Ancien Regime armies with conscripted citizens on a scale far greater than the traditional Prussian canton system. French corps operated semi-independently, foraging off the land, moving faster than a traditional supply train could follow. Commanders were selected by merit and ruthlessly replaced when they failed. The French Republic was forging fundamentally different methods of modern war.

The Prussian high command watched all of this and shrugged chauvinistically. Even after the Prussian defeat at Valmy in 1792, France’s further successes against Austria and the smaller German states were attributed to luck, revolutionary fervor (a temporary condition, surely), and the perceived incompetence of Prussia’s Habsburg rivals. The institutional conclusions that should have followed were systematically suppressed by an officer corps whose identities, social status, and economic livelihoods were intertwined with the existing system. Admitting that the system was broken meant admitting that they were, too.

The Outsider’s Audit

Scharnhorst understood the Prussian problem because he was not quite one of them. Born in 1755 to a family of modest Hanoverian freeholders he had no aristocratic sinecure to protect. He had earned every position through demonstrated intellectual and professional ability, which made him unusual in every army in which he served. He joined the Hanoverian artillery, studied obsessively, and began publishing his own military journal in the late 1770s when he was barely in his twenties. By the time he was a first lieutenant, he had already written more seriously about the theory of war than most generals had read. But Scharnhorst was no desk jockey. When the wars of the French Revolution erupted in 1789, he joined the Hanoverian Expeditionary Corps under Prince Frederick, Duke of York and fought with great distinction in the Netherlands.

Prussia’s Frederick William III recognized something in Scharnhorst. In 1797, Prussia invited him to the kingdom with an offer of a promotion to lieutenant colonel. Scharnhorst demurred, preferring to stay in Hannover with his writing and his wife Clara. In 1801, the King of Prussia invited him again to Berlin with the promise of a nobility patent (making him Gerhard ‘von’ Scharnhorst), a double salary, and a promotion to colonel.

This time Scharnhorst accepted and moved with his wife to Berlin, where the king appointed him to lead the newly founded Akademie für junge Offiziere der Infanterie und Kavallerie. It was an experiment in building what we would now call institutional knowledge. Scharnhorst occupied a townhouse on the prestigious Unter den Linden boulevard and put out a request for regiments to send him their best and brightest young officers to teach them the art of war. He meant to expose them to that which the Prussian military had rarely prioritized: serious, critical thought about why armies won and lost, not just how they maneuvered and fought. He also started a seminar known as the Militärische Gesellschaft or “Military Society” for similarly aligned officers such as the brilliant August von Gneisenau.

The old guard was not pleased. To men like Field Marshall Wichard von Möllendorf, the Akademie looked very suspicious. A boondoggle. Officers who asked too many questions were dangerous. Officers who read theory were practically treasonous. The aristocratic establishment that ran Prussia’s military viewed Scharnhorst’s project as an irritant at best. At worst, they feared it was a seditious little intellectual club that—even if it wasn’t fomenting republicanism—would certainly produce nothing useful for the real business of war.

The Prussian general of the day would have heartily agreed with Britain’s Prince George, Duke of Cambridge, when he asserted: “Brains? I don’t believe in brains! You haven’t any, I know, sir! There is a time for everything, and the time for change is when you can no longer help it.”

So Möllendorf and the other generals kept Scharnhorst far away from any influence over doctrine, procurement, or command selection. Only a handful of young ensigns and cornets were sent to the Akademie, largely outcasts that the colonels didn’t mind losing to their young king’s eccentric project.

Let the Hanoverian Colonel Scharnhorst have his cadets and run his Francophilic seminar. He could not actually change anything that mattered.

The Star Student No One Wanted

One of the young officers Scharnhorst received was a twenty-one-year-old lieutenant from Prince Ferdinand’s 34th Infantry Regiment named Carl von Clausewitz—a young man who, by almost every conventional Prussian measure, had no business being in the room.

Lieutenant Clausewitz was from Magdeburg. His father was a minor civil servant and former officer, technically of noble birth, but he had retired from the army at the same time as Frederick the Great’s purge of non-noble officers. Clausewitz’s tenuous claim to his “von” (along with his suspiciously Slavic-sounding surname) led to loud questions from officers with family lines stretching back before the foundation of Prussia.

To make matters worse, Clausewitz was socially awkward, bookish, and possessed of a neurodivergent intensity that made him deeply unsuited for both small talk and the glad-handing politics of the regimental mess. He did not mesh well with the kind of men who ran things. He had earned a battlefield commission during the Siege of Mainz in 1794. He had no patrons, no family connections to command, no path to preferment.

What he did have was an extraordinary mind.

Scharnhorst saw it immediately. He took Clausewitz under his wing, guided his reading, sharpened his thinking, and (critically) advocated for him in the face of a system that would have been happy to leave him in obscurity. When Clausewitz’s noble credentials were challenged in 1803, Scharnhorst intervened with the king to arrange for Clausewitz to be made aide-de-camp to Prince August. He used personal capital that a more self-interested man would have husbanded carefully. He spent it on a young man his colleagues considered more trouble than he was worth.

Scharnhorst had grasped something the establishment was incapable of grasping: that the asset which would determine Prussia’s future military effectiveness was not the bayonet or the battalion square, but the caliber of mind that could understand war well enough to adapt to it in real time. He was building intellectual infrastructure.

The institution, for the moment, was entirely uninterested.

The Prussian Blue Screen of Death

On October 14, 1806, the parade ground army met the real thing at the twin battles of Jena and Auerstedt.

In a single day, Napoleon’s forces completely destroyed the Prussian field army of 70,000 men. Prussian formations that had drilled for years dissolved. Commanders who had been promoted for lineage rather than ability fought with tremendous bravery and failed in ways that were, in retrospect, entirely predictable. The rigidity that the old guard had mistaken for discipline became a death sentence when the French corps commander Louis-Nicolas Davout improvised, flanked, and refused to fight on Prussian terms.

The commander of the Prussian main body, the Duke of Brunswick, was shot in the head while trying to rally his grenadiers and killed. As the main body fell back in disarray, they ran smack into the Prussian vanguard, which was fleeing north as fast as they could from Napoleon and four other French corps.

Within three weeks, the Kingdom of Prussia was prostrate. French troops occupied Berlin. The king fled east to Königsberg. The army that had been the terror of Europe for sixty years had been broken faster than anyone thought possible in a humiliation greater even than Austerlitz.

Scharnhorst, who had argued for years that a reckoning like this was coming, was given command of the rear guard and was wounded in the subsequent holding action to buy time for the king to escape. He was then assigned to manage the armistice negotiations.

Lieutenant Clausewitz spent a year as a prisoner of war before being exchanged.

The Seeds and the Sower

Here is the part of the story that matters most: Scharnhorst’s post-Jena reforms were not built from scratch.

Scharnhorst’s Militärische Gesellschaft had produced a cohort of officers that understood that Prussia’s loss was not a question of bravery or morale. It was the result of a structural failure. These men had read military history, studied Napoleon’s campaigns, and learned to ask hard questions rather than reflexively defer to seniority. Scharnhorst now had something to work with. The King of Prussia made him a general and gave him carte blanche to reform the army.

The reforms Scharnhorst rammed through between 1807 and his death in 1813 were sweeping: the abolition of corporal punishment; the transformation of the Akademie für junge Offiziere into the famous Prussian Kriegsakademie, the war college that would produce professionally educated officers for generations; the opening of commissions to non-nobles on the basis of merit; and the development of the Landwehr, a reserve system that could mobilize the nation’s manpower beyond just the standing army. He also pushed for the development of the general staff as an institutional brain trust rather than a collection of royal favorites—a concept so foreign that it required the complete destruction of the old system to implement.

Major Clausewitz completed the transformation, becoming the superintendent of the Kriegsakademie from 1818 to 1830. The thinking he had done under Scharnhorst’s guidance was refined over years of study and tested against the catastrophe of 1806. Clausewitz, with the extensive help of his loving and very patient wife Marie, produced On War, the most important work of strategic theory in the Western canon.

The Prussian army that defeated Napoleon at Waterloo in 1815 and the Prussian state that became the dominant power in Europe by 1871 (along with the modern United States officer corps) stood on the foundation Scharnhorst laid before anyone thought a foundation was needed.

What This Means for Today

Scharnhorst’s life is a story about the cost of institutional complacency and the mechanism by which that cost eventually gets paid.

The Prussian officer of 1805 was not stupid. He was not a coward. He was a rational actor operating within a system that rewarded the perpetuation of existing structures over honest assessment of emerging threats. The incentives ran in exactly the wrong direction. The intelligence was there. The information was available. The analytical capacity existed—somewhere—in the system. But the institutional architecture was built to filter it out.

Scharnhorst’s answer was to build a parallel infrastructure—a small, underfunded, politically marginal military academy—and populate it with aligned men who had no vested interest in defending the existing answers. He could not reform the institution from the inside through persuasion; it was too well-defended for that. He built the capability that the institution would need when it finally had no choice but to change.

Armies, governments, and industries have always preferred the comfortable illusion that existing structures can be changed when needed over the expensive, politically fraught work of building what is needed in advance. The Prussian high command of 1805 was certain that its machine would perform when called upon. Its failure was comprehensive, rapid, and very nearly fatal.

The problem that the West faces today is a version of the same. The systems, processes, and procurement cultures built for a post-Cold War world of limited contingencies and leisurely acquisition cycles are not suited to the current environment. The gap between what existing architecture can deliver and what the moment requires is widening. The Jenas of the next decade will not announce themselves in advance.

Scharnhorst did not wait for Jena to start building the Kriegsakademie. He started five years before with a handful of students and a small budget.

That is the lesson. Not the dramatic rebuild after the disaster, but quiet, stubborn investment in the face of institutional intransigence.

The Prussian Army eventually got its reformation, but the cost of waiting was severe and entirely avoidable.

Gerhard von Scharnhorst knew that. He said so, repeatedly, before anyone cared to hear it.

That’s the sort of vision after which warships are named.

Think tank reports rarely make for compelling reading. That’s why when we come across one that defies the genre, we have no choice but to recommend it. The Foundation for American Innovation (FAI) recently published Going on Wartime Footing: Five Big Reforms for American Defense. Author Tim Hwang defines wartime footing as “a wholesale mobilization of the defense industrial base that would prepare the country for major combat operations and the prospect of a scale of conflict not seen for a generation.” As the authors of a recent book about this very topic, we agree any reforms need to match the scale of our current emergency.

From a Citizen’s Proving Ground to a Defense Acquisition Delta Force, Hwang’s proposals are ambitious. They also exist in the realm of the possible. We’re reminded of William Greenwalt’s and Dan Patt’s bold call to abolish JCIDS, the joint requirements process. It probably seemed outlandish to some, but it came to fruition just six months later under Secretary Hegseth. And like the simple proposal to kill the joint requirements process, Hwang’s reforms are legible to the lay person because they address the root causes of the dysfunction. There is no obfuscation by “experts” who propose tinkering with arcane processes at the margins.

Hwang’s reforms, while creative, are not without precedent. They may seem farfetched or too politically fraught to those who are accustomed to incremental adjustments and inaction. But history suggests that we can—and should—think bigger.

Big, Beautiful Reforms

The defense acquisition bureaucracy employs nearly 160,000 individuals across civilian and uniformed personnel. That’s about the size of the U.S. Marine Corps. What do we have to show for this shadow service? A weapons arsenal largely designed during the Cold War. While some think tanks argue that we just need more acquisition personnel, FAI argues for finding the right people via a “Defense Acquisitions Delta Force.” It may be less exciting than the real Delta Force, but it would be no less important. This “crack team of elite military officers whose purpose is to serve as an acquisitions special operations squad . . . would be tasked with the hardest defense procurement missions.” The chosen few would be up against a massive bureaucracy where inertia is the default. Success would be challenging, but possible.

America had a version of the Defense Acquisitions Delta in the late 1970s under Bill Perry, under secretary of defense for research and engineering. Perry and his small team—a delta force in spirit, if not in name—moved with purpose to acquire the capabilities for the Second Offset strategy. Within five years, stealth, GPS, and precision-guided munitions were all operational. Showing the spirit of a special operator, Perry did whatever it took to win. He circumvented the traditional and slow PPBE acquisition process, killed programs that weren’t delivering, and forced through the efforts that were. That he had no formal acquisition background was almost certainly a boon. If the Department of War (DoW) brings in Perry-quality talent, the Defense Acquisitions Delta has a real shot at success.

Adjacent to the massive acquisition bureaucracy are the resource-intensive government labs, warfare centers, FFRDCs, and more. These DoW-supported institutions compete directly with private industry. For Hwang, it’s high time we “eliminate the shadow competition.” The government competing with industry is not a new problem, just one that’s been remarkably resilient to change. Writing in 1995, Jacques Gansler, who would go on to be President Clinton’s under secretary of defense for acquisition and sustainment, complained of the conflict of interest where the government is “both a direct competitor and the possessor of the right to decide who wins and loses.”1 The dynamic is all the more egregious because the robust private sector “offers essentially identical services.”2 Aside from truly unique areas, like nuclear weapons, Gansler generally opposed the duplicative existence of labs and FFRDCs.

Hwang finds the shadow competition least excusable when it comes to shipbuilding. The Naval Sea Command (NAVSEA) is responsible for acquiring ships and combat systems. In many cases, it also architects and builds those same vessels via its government labs and shipyards, disadvantaging a commercial industry hungry for work. Perhaps the only thing more shocking than learning China has over 200 times more shipbuilding capacity than the United States is learning that NAVSEA has nearly 90,000 employees.

The shunting aside of the commercial shipbuilding industry recalls the trials and tribulations of Andrew Higgins. The outsider boatbuilder from Louisiana built over 90% of all vessels in the U.S. fleet during World War II. But first, Higgins had to go head-to-head to compete with the Navy’s Bureau of Ships (NAVSEA’s predecessor). As late as 1942, the Bureau of Ships blocked him from competition and stole his designs. It was only when Senator Harry S. Truman directly intervened that Higgins finally got the contracts he deserved. Higgins’ assessment of the shadow competition rings true today: “Nothing could be healthier for the Navy as a whole, and the country that they really desire to serve, than that there be a ‘house cleaning’ in the Bureau of Ships. It would be preferable that quite a number of officers and civilians therein go to other duties.”3 Higgins understood then, as Hwang does now, that commercial shipbuilding is a national imperative.

As a talented civilian innovator, Higgins would have been a fan of another of Hwang’s recommendations: a citizen’s proving ground. The DoW would operate a “nationwide network of munitions proving grounds on underutilized federal land, military training areas, or testing ranges.” After obtaining a license, citizens would then be able to conduct weapons and munitions testing, “particularly against defined priorities and targets set by the Pentagon.” It sounds crazy but is entirely possible and aligned with the American innovator spirit. A citizen’s proving ground is a necessary response to the democratization of weapons development via commodity hardware and commercial software and AI. As we’ve seen in Ukraine and Iran, cheap drones are now a fact of modern warfare. Critical innovation happens on a daily or weekly basis—not according to years-long procurement cycles.

While the barriers to build weapons fell, the United States doubled down on the institutionalization of defense technologies. This yielded formidable advances like stealth and intercontinental ballistic missiles. It also crowded out the tinkerers and cowboys who have historically played an important role in developing new tech. Early American rocketry was defined by Robbert Goddard and the Caltech Suicide Squad. Goddard launched the first liquid-propellant rocket at his aunt’s cabbage farm in Massachusetts in 1926; when he could no longer ignore Worcester County’s disgruntled residents, he relocated to spacious New Mexico to continue his experiments. Now NASA’s premier spaceflight center is named after him.

Meanwhile, the Suicide Squad, an inexperienced group of graduate students, were busy testing rocket motors in a “semi-official manner” at Caltech. They operated with no budget and a regard for safety typical of college kids. After they were kicked out of the campus laboratory following an incident involving corrosive nitrogen tetroxide, they moved a few miles to the Arroyo Seco canyon to continue making mayhem. In 1939, General Hap Arnold, commander of the Army Air Forces, paid the group a visit. He liked what he saw. The Suicide Squad promptly received an Army contract for the first jet-assisted takeoff rockets in 1939. By 1943, the group formally became known as the Jet Propulsion Laboratory (JPL). Today’s buttoned-up JPL belies its rag-tag origins. The early aviation and firearm industries come from similar traditions. We would do well to tap the exceptional—if unusual—talent that can be found far outside the Beltway and national labs.

Hwang provides the historical precedent for his remaining two proposals. There’s “The Billion Dollar OTA,” which would be singularly focused on “whether a company is able to deliver systems meeting defined operational metrics. If a contractor is able to meet these requirements, the government would guarantee a certain purchase volume worth billions.” The United States had such an outcomes-based budget for shipbuilding during World War II. Henry Kaiser had the freedom to innovate on mass production for his Liberty Ships because the government defined what ships it needed and committed to buying them in huge quantities. If government provides a clear demand signal and leaves innovation to industry, we can approximate the healthy dynamics of a commercial market.

The final proposal is the simplest: “The Secretary’s Firing Line.” It’s a bet on people over process. The Secretary of War would “bring together the entirety of top military leadership to directly review all legacy programs and cut ruthlessly where needed.” This is something Army Chief of Staff George Marshall implemented to great effect during World War II. He conducted regular program reviews that forced staff to rank programs and defend them. One of the results was a shift away from tank destroyer units and an expansion of armored divisions and artillery. The firing line is a highly leveraged concept the Secretary can implement today.

Every one of Hwang’s proposals has historical precedent. We know they are feasible. Now we need to discover the will to act.

Arthur Herman is the Pulitzer Prize Finalist author of Freedom’s Forge: How American Business Produced Victory in World War II. His newest book, Founder’s Fire: From 1776 to the Age of Trump, will be released by Hachette/Center Street on April 21. Find him on X at @ArthurLHerman

It’s a question historians have often asked themselves. It’s one Nazis caught in the Führerbunker, and later on trial at Nurenberg, must have pondered, as well.

What went wrong, they wonder, when Nazi Germany had so many trump cards in its hand?

A world-class industrial economy.

A first-rate military. (“The most professionally skillful army of modern times,” according to Martin van Creveld, one that, according to Colonel Trevor Dupuy’s analysis, was able to inflict 50 percent more casualties on its British and American opponents under all conditions than its opponents were able to inflict on it).1

An exquisite community of brilliantly trained scientists and engineers.

Above all, a fistful of advanced—even futuristic—military technologies. These included the first jet fighters and aircraft, the first killer drones and ballistic missiles, the first precision-guided munitions (the Fritz-X radio-guided bomb), the world’s most advanced submarine design, and an atomic bomb program underway by 1939.

Imagine what might have happened if Germany had managed to deploy its jet aircraft during the Battle of Britain. Imagine if the V-2 rockets had been improved to reach Moscow and even New York City. Imagine Germany using the first atomic bomb to stop the D-Day invasion, as many in the Manhattan Project had feared.

Yet the truth is, Germany lacked what the United States had and was able to bring to the other Allies: an industrial base large and flexible enough to scale production of the military technologies that did give the Allies their decisive edge: tanks, trucks, conventional fighters and bombers, artillery pieces and machine guns, submarines and destroyers and aircraft carriers—by 1944 eight a month—as well as the freighters needed to carry all this material to battlefields across two oceans.2

The rest—including the atomic bomb—was largely icing on a massively conventional cake.

That’s not to say innovation didn’t play its part in victory. The weapons the Allies started the war with—with the exception of the Supermarine Spitfire fighter—were markedly inferior to those used by the Axis. By the last year of the war, however, America and its allies outclassed the Axis in every area that counted, from tanks to fighters to bombers to artillery fuses.

That truth reveals a larger point: innovative military technologies don’t change the larger strategic balance. They express the larger strategic balance. In that sense, Germany was beaten two years before the war actually ended. By the summer of 1943 the Allies had taken Sicily and the Soviets had crushed the Wehrmacht’s last offensive in the East at Kursk—while U-boat losses forced the German navy to halt its campaign to control the Atlantic at the end of May.

People began avoiding giving the Nazi salute whenever they could. The German security service reported that no senior industry leaders still believed the war was winnable.3

Yet the war itself dragged on until May 1945. What followed was simply a regime struggling to escape its inevitable fate: downfall and defeat.

******

So went wrong for Nazi Germany?

Let’s start with its industrial base, the third largest in the world in 1939. (Tied for second with the Soviet Union if Austria and Czechoslovakia are included).

Further: with the fall of France and occupation of Holland, Belgium, Denmark, and Norway, Nazi Germany had captured the industrial heart of Europe. Hitler controlled an economic bloc equal to the GDP of the United States, when you include allies like fascist Italy—even excluding Germany’s potential ally and biggest trading partner in 1940-1, the Soviet Union. Germany’s sphere of influence at its zenith covered one-fifth of the world’s population: or roughly that of the entire British Empire.4

Yet even with this formidable base, the German war machine lacked two essential ingredients. The first was energy, specifically oil, while the United States was able to supply its allies and meet its own needs and more throughout the war. 5 Access to sufficient oil to keep its armies, navy, and air force running was Germany’s Achilles heel, right from the start.

The other was workforce. Here Germany’s first-rate military demanded more men than the nation’s demographics could support, especially after the invasion of the Soviet Union. Incredibly, by the fall of 1941 virtually every German male in his twenties had already been called up.6 Every demand for more men afterward to fight a three-front war (Western Europe, Russia, and North Africa/Italy) increasingly drained manpower from factories and shipyards; the only alternative for keeping the industrial base going was slave labor. By 1944 more than one-third of workers in the German war industries were unwilling foreigners, including Jews destined for the gas chamber. Even in 1943, 80 percent of the workforce making Stuka bombers were Russians.7

Foreign workers, of course, were not as efficient as German counterparts. Then, even when a superhuman effort was able to surge the production of fighters in the desperate months of 1944, still there was no fuel for flying them.

What about the much-vaunted secret weapons?

We can start with jet propulsion. Heinkel first tested a jet-powered prototype in August 1939, a month before war broke out. Both Heinkel and Messerschmidt immediately plunged into developing jet aircraft. So swift was their progress that some worried the entire prop-driven aircraft development program might be in jeopardy.8 Hitler gave his enthusiastic backing to Messerschmidt’s first designs in the summer of 1942. By the end of May 1943, the Air Ministry was pushing Messerschmidt to start mass production.

If Messerschmidt had somehow succeeded, the entire U.S. daylight bombing campaign that fall might have been swept from the skies by squadrons of Me 262’s. But Messerschmidt couldn’t do it. Going from prototype to production models takes thousands of hours of testing and evaluation, then months of experimental series production. Messerschmidt had no AI or computers or other information technology to accelerate the process; even the Junkers-Jumo jet engine the fighter required wasn’t ready for limited series production until the summer of 1944, when the Allied bombing campaign was approaching its peak.

Instead, it was a prop-driven fighter plane, the P-51 Mustang, which had shifted the balance in the air war over Germany and Europe, escorting thousands of prop-driven bombers to their targets. The Eighth Air Force’s fighter strength quadrupled in just eight months in 1944, while German air losses that summer amounted to half its aircraft and one-quarter of its pilots every month.9

When the 262 finally did take to the air in September, it was still largely in a trial and development unit. Its top pilots died all the time from accidents, while its slow landing speed made it an easy target for P-51’s roaming the skies at will.

Despite claims by German air ace Adolf Galland and others that it was Hitler’s obsession with using the 262 as a fighter-bomber instead of an interceptor that kept it from being a decisive weapon, what actually kept the German jet program from tipping the balance in the air war “was not incompetence or conservatism but the debilitating material limitations of the German war economy”—limitations that did not constrain America’s industrial base.10

Something similar happened with Germany’s rocket program. Here again the Germans were off to a flying start, as it were, going back to 1923 when a German scientist published The Rocket into Interplanetary Space. The German army set up its first top-secret rocket laboratory in 1932, just as Hitler was becoming chancellor. Its first employee was a 20-year-old student, Wernher von Braun.11

The funding for testing and development in 1935 was generous enough to enable von Braun and colleagues to set up a launch pad and supporting lab and production facilities at Peenemünde in northern Germany. Still, it took six years of relentless testing and serial failures before the team had a rocket prototype with which to approach Adolf Hitler in August 1941.

The Führer became rocketry’s enthusiastic champion and hailed the new missile as a military revolution. But it would be another year before the rocket had its first successful operational flight in October 1942, just before American and British troops landed in North Africa in Operation Torch. Nonetheless, Hitler dubbed the rocket “the decisive weapon of the war,” and ordered five thousand for immediate production.

That was a fantasy number. In fact, technical difficulties held up production repeatedly. Then, when the Luftwaffe offered its own “flying bomb,” the so-called V-1, resources were diverted to bring it into production first. It wasn’t until July 1943 that it seemed von Braun finally had the missile he and Hitler wanted: but an RAF air raid the next month virtually destroyed the Peenemünde site, setting production back months.

Now the SS stepped in, taking over production of the V-2 (as it was now designated) in a brand-new site deep underground in the Harz mountains, using slave labor to build the site and then construct the rockets. By January 1944, production finally got underway with an initial contract for 12,000. That month SS managers produced exactly three V-2s, all of which had serious production defects.12 It took another eight months before the first V-2 was fired at its principal target, London. Others were launched almost daily at Antwerp, where the Allies were building up for their final offensive into Germany.

In the end, six thousand V-2’s were built—half the promised contract—and 1,403 were launched, with just 517 landing on or near their target. Unlike the V-1, the supersonic V-2 arrived virtually without warning, with a huge field of destruction.

Together with the V-1—of which some 2,420 landed on target out of a total production of 30,000—the V-2 earned its reputation as Hitler’s “terror weapon.” Together they killed some 9,000 British civilians.

Still, it’s instructive to realize that the 2,500 tons of explosives both weapons delivered over a nine-month campaign equaled less than a quarter percent of the tonnage the Americans and RAF unloaded on Germany during the same period. 13 That destructive power depended not on advanced space-age technologies, or even innovative scientific breakthroughs, but a grim dedication to putting massive amounts of steel on target, while Allied air forces decimated the Luftwaffe and Allied armies swept back the Wehrmacht month after month.

Whether that Allied bombing campaign had a strategic impact on the war worthy of the cost in materiel and lives has been debated almost from the moment the war ended. What we can say is that the Germans themselves knew that while the bombing did not and could not halt their wartime production, it could severely curtail its potential. When German industrial leaders met in January 1945, they discovered that in 1944 they had produced 35 precent fewer tanks, 31 percent fewer planes, and 42 percent fewer military trucks than planned—almost entirely due to Allied bombing.14

By contrast, during that same year America’s factories and shipyards were producing a warplane every five minutes, eight aircraft carriers a month, fifty merchant ships a day, and 150 tons of steel every minute—whereas Germany was lucky to get a third of America’s industrial output that same year. 15

Sclerotic production also doomed Germany’s efforts to build its advanced submarine, the Type XXI, of which 118 were commissioned but only one actually carried out a combat patrol before the war ended.16

Likewise, Germany’s radio-guided munition, the so-called Fritz X, did severe damage to the handful of ships it struck in the Mediterranean when it was deployed in 1943 (including, ironically, an Italian battleship), but then faded from use when the Allies learned how easy it was to disrupt its radio guidance using an early version of electronic warfare.17

Which brings us to the ultimate weapon: the atomic bomb.

One disadvantage of the German nuclear program was Hitler’s lack of interest. For Hitler and other Nazis, atomic physics was still “Jewish science,” even after most of those Jewish scientists like Albert Einstein had left Germany. Hitler never attempted to stop the program, but he never gave it the funding or passionate attention of other secret weapons.

One big advantage was that many of the leading scientists who had created nuclear physics were still working for the Reich, and most were involved in the atomic bomb project from its start in 1939. That year, chemist Otto Hahn published the first paper demonstrating that uranium was capable of nuclear fission; German physicists soon realized a fission reaction would produce an explosion of almost unimaginable power. Another chemist and member of the Nazi Party, Paul Harteck, alerted the Wehrmacht of this potential, and by September—even as German tanks were rolling into Poland—the army had set up its first research team and lab. In December, Germany’s star nuclear physicist, Werner Heisenberg, explained in detail how such a weapon could be created.

Unfortunately for the Nazis, and very fortunately for the Allies, the German scientists chose the most complicated and circuitous path imaginable to create the U-235 needed for fission. They decided using plutonium as a cheaper substitute was impractical (plutonium would be the basis for the bomb the Americans would drop on Nagasaki) and opted for using intermediate materials to slow the nuclear fission reaction of U-235 down to controllable levels. One of those materials was graphite, but the German aircraft industry needed all the graphite it could get, so that wasn’t an option. Instead, the German program had to rely on deuterium or “heavy water,” which had only one source: a hydroelectric plant in far-off Norway.

Once the Allies caught on to what the Germans were doing—even as their own joint U.S.-British-Canadian atomic bomb project was surging ahead—they sent commando teams to disrupt the production and outflow of the heavy water from Vemork in Norway.18 In the end, the Germans never produced more than half the heavy water the program needed; a disappointed Wehrmacht passed the program over to the Education Ministry in 1942, a sure sign no one took the building of an atomic bomb very seriously. When Allied investigators examined the remains of the program in 1945, they found Germany were still years away from building anything like the weapons which America had used to end the war in Japan.

The Germans had also seriously underestimated the amount of U-235 needed to create a nuclear reaction leading to an explosion, thanks to Heisenberg’s miscalculations (whether he did so accidentally or deliberately, in order to undermine the Nazi effort, is still not clear). The larger point is that the German program was entirely dominated by scientists rather than engineers: scientists who were more interested in proving theory than devising workable solutions. When they ran into technical difficulties, as they often did, the program stalled and hesitated.

By contrast, the Allied bomb project found all the engineers it needed through the private companies who were intimately involved in every stage of the program, like DuPont, Union Carbide, Monsanto, Chrysler, Stone & Webster, and many others. America and Britain never hesitated to commit its best minds but also its best companies to the project, knowing they would find the solutions to whatever obstacles the program faced.19 And when the British realized it was a project beyond their industrial resources, they quickly handed it over to the Americans and American industry.

No such hand-off was possible with the German program. Instead, years were wasted in large part because the resources were never there, material or intellectual, to achieve a final result. Even the greatest industrial power in the world, devoting 150,000 personnel and $2 billion to the Manhattan Project ($50 billion in today’s dollars), could not manage to create a workable bomb until after Germany had been defeated.

With the atomic bomb, like so many other secret weapons programs, the Germans were in a race to nowhere. The simple fact is that the United States and the allies had the resources to devote to victory. Germany did not.

****

What are the lessons for today?

The first is worth repeating: new defense technologies, no matter how advanced, don’t change the strategic balance. They express the strategic balance, reflecting the material resources in men, machines, and production that created that balance in the first place. Even the atomic bomb was used when the war was already won in Europe and effectively won in the Pacific. By the time Germany’s jet fighters and V-2 rockets became operational, Germany was all but defeated. It was Allied armies and industries, not secret weapons, that determined the tide of war.

The second lesson resonates with today’s defense industrial contest with China and Russia, and even with the current war against Iran.

It’s often said that war is a contest of wills, and the object of victory is to destroy the opponent’s will to fight. That applies to cold wars as much as hot, kinetic ones.

It’s important to acknowledge that China has been on an industrial production “war footing” resembling the United States in World War Two for several years now, while the United States itself has more closely resembled Nazi Germany in its obsession with exquisite and expensive weapons systems like the F-35 and Gerald Ford-class carriers, and an equally exquisite industrial base focused on quality at the expense of quantity.20

Still, if dire warnings about China’s proficiency in technology and defense production serve to discourage and dismay Americans instead of encouraging us to urgent action, then the fault lies not with Beijing but closer to home. The American industrial base that produced victory in World War II no longer exists, but a new industrial landscape is developing before our eyes.21 With it produce a host of new defense companies that can compete with the fabled Big Six--,Lockheed Martin, RTX (formerly Raytheon Technologies), Northrop Grumman, Boeing, General Dynamics, and BAE Systems Inc.--ones that bring those new technologies like AI and autonomy directly in line with productive resources and capabilities.22

And that’s the key point. In the end, what defeated Germany weren’t better weapons or even more industrial output, but the ability to bring the two together in an ever-ascending spiral of innovation and production. No other country, not even China, is able to match the promise the United States has of doing it again.

The crucial question now is, are we poised to end up like Germany during World War II, stuck with lots of great innovative weapons but no means of scaling them into action? Or will we end up like our parents and grandparents in the United States, welding innovation and productivity together into an invincible combination?

The answer will determine the future of the 21^st century.

Lucas Cristaldi is an incoming M.S. Computer Engineering student at New York University and a former U.S. Air Force Contracting Officer who executed acquisition operations across CENTCOM, including contracts supporting the Afghanistan retrograde and Agile Combat Employment exercises in Palau and Puerto Rico. He co-founded the Business Innovation Cell at Moody Air Force Base’s 23rd Contracting Squadron. Lucas is a graduate of Palantir’s American Tech Fellowship.

Matt Brien works for a defense technology company in California and is a Reservist in the U.S. Air Force. Matt’s experience as a contracting officer includes operational contracting support, Director of Business Operations at the 386th Expeditionary Contracting Squadron, USSOCOM Ghost #321, and complex systems procurement at Space Systems Command.

Take a closer look at the everyday items on your desk or at home. Flip over your keyboard, your mouse, or your phone case, or examine the government award you proudly display. You will likely find the words “Made in China” stamped on the back.

China’s industrial base accounts for about 29% of global manufacturing output. China dominates critical dual-use sectors, from shipbuilding (53% of global production) to commercial drones (over 90% of the global market). Under Beijing’s Military-Civil Fusion strategy, this civilian industrial capacity can be rapidly redirected toward military production. The U.S. manufacturing sector, meanwhile, has contracted to just 8% of American jobs compared to 32% in 1953.

These aren’t distant problems from our daily lives. Every “Buy Now” click potentially strengthens our adversary’s manufacturing base and indirectly supports China. The rise of online shopping has obscured country-of-origin information, making many Americans unwitting financial supporters of rival industrial capacity.

The crisis extends beyond individual purchases to institutional procurement. The Defense Acquisition Workforce oversees massive spending flows that could either strengthen American industrial capacity or accelerate its decline. The cumulative effect of routine purchases—Micro-Purchase Threshold (MPT) transactions, Operation and Maintenance (O&M) spending, and Government Purchase Card (GPC) transactions—represents billions in economic activity that could support domestic manufacturers or foreign competitors.

This is an opportunity. While major weapons systems require complex international supply chains, routine institutional purchases offer immediate opportunities to build domestic capacity. Air Combat Command installations alone spent hundreds of millions on GPC transactions. Even small percentages flowing to foreign manufacturers represent substantial sums that could support American businesses instead.

We know these flows can be redirected fast because we tackled the problem as part of a contracting squadron in Valdosta, Georgia. By expanding local vendor pools, partnering with a technology startup to surface American-made alternatives, and building direct relationships between a military installation and its surrounding community, we kept federal dollars domestic and proved that the model can be replicated at installations across the country.

More than any policy shift, what’s needed is a change in mindset. Rebuilding American industrial resilience requires acquisition professionals who refuse to be spectators.

‘It Won’t Fail Because of Me’

In October 2021, while strategists debated supply chain vulnerabilities in conference rooms, we decided to start where we stood. As Contracting Officers with the 23rd Contracting Squadron at Moody Air Force Base in Valdosta, Georgia, our day-to-day work was buying goods and services and executing contracts for the installation. As part of that work, we decided to address foreign dependency in defense procurement in our own, small way.

We started a Business Innovation Cell, supported by leadership that understood that Airmen with time, resources, and permission to solve problems could generate solutions with lasting impact. We saw a glaring gap: a disconnect between what was available in the local market and how the government was buying. Procurement processes were stuck in decades-old rules, not actively taking advantage of capable companies that could deliver quicker, better, and closer to home. Our mandate was to close that gap, to engage local small businesses, keep federal dollars within America, and strengthen the distributed network of enterprises that provides actual resilience during crises. We did this by visiting businesses directly, educating them on government contracting, and actively expanding the pool of qualified vendors for routine purchases.

Defense acquisition professionals can easily fall into a rut, viewing themselves as passive processors of requirements. We decided to live out a different philosophy: “It won’t fail because of me.” This was the operational philosophy behind humanity’s greatest achievements and that helped put Americans on the moon. During Apollo, 400,000 Americans from different backgrounds shared one common commitment: personal accountability for collective success. Each person understood that mission success wasn’t someone else’s problem; it was everyone’s responsibility to ensure their piece worked flawlessly.

We applied this same mindset to fixing defense acquisition from the ground up. Rather than becoming spectators waiting for someone else to solve industrial base vulnerabilities, we recognized we were part of a community we could actively shape.

Our first intervention was symbolic but telling. When rescuemen at Moody Air Force Base—Airmen who risked their lives to save others—received commendation awards with “Made in China” stickers, we recognized an opportunity to act. Heroes being honored with foreign-manufactured awards represented a systemic failure, and we knew that the solution required no complex bureaucratic reform. Numerous small businesses in the Valdosta area could do the job. Many could do it better.

We sourced the awards from local vendors in the Valdosta community and helped on-ramp those small businesses as qualified suppliers to the base. The new awards were crafted by local artisans who understood the significance of what they were creating and took personal pride in honoring American heroes properly, instead of being mass produced on some generic, far-away assembly line. This was a small victory, but it proved that in many cases the solution was already in our backyard. So what else were we missing?

Another operational challenge emerged with off-road vehicle maintenance. Moody AFB struggled with ATV repairs, facing lengthy delays through traditional contractors. Yet driving through Valdosta revealed an obvious solution: local businesses specializing in everything from golf carts to UTVs to motorcycles. Rather than waiting for vendors to discover government opportunities, we redefined our approach to expand the competitive pool. Federal guidance emphasizes that contracting officers should actively encourage small business participation. Therefore, we hunted for qualified local capabilities, visiting and educating potential vendors about opportunities.

By expanding the vendor pool from the usual regional contractors to 50+ qualified local businesses, we created genuine market competition that drove down costs, improved delivery, and enhanced responsiveness. Every contract still went through the competitive bidding process, but now we had more qualified bidders competing for the work.

We also partnered with a technology startup that had developed a browser extension and product-tagging platform designed to identify the country of origin for products on Amazon and surface American-made alternatives. Through this partnership, we worked to integrate product-origin data into our procurement workflow, so that when cardholders shopped for supplies online, they could immediately see where items were manufactured and find domestic substitutes.

The downstream effects of these simple reforms extended throughout Valdosta’s manufacturing ecosystem. When federal dollars stayed local, they created multiplier effects: construction contractors sourced materials from Valdosta’s glass manufacturers, roofing suppliers, and fabricated metals producers. We observed increased participation from local vendors in subsequent contracting cycles, and our relationship with Lowndes County government deepened as shared economic interest aligned the base and the community.

The February 2023 Industry Day at Moody AFB represented the culmination of these efforts. Unlike standard industry events, this gathering focused on local small business contractors in three areas: services, construction, and supplies. It united wing leadership, community entrepreneurs, and government officials. Moody had spent millions in South Georgia, but beyond dollar figures, we achieved community unity around shared prosperity.

A Replicable Model

These examples show that routine O&M purchases can provide vital revenue streams that strengthen the distributed manufacturing network America needs for strategic resilience. It all depends on millions of daily, individual choices—and ultimately, on the mindset that acquisition professionals bring to the work.

Are we passive requirement processors or active contributors to national resilience? Do we accept foreign dependency as inevitable, or actively seek American alternatives? Every “Submit Order” click represents a choice: strengthen America’s industrial base or fund our adversaries’ modernization efforts.

The path forward requires acquisition professionals who understand that their individual decisions, multiplied across thousands of installations and billions in spending, shape America’s strategic resilience. It requires rejecting spectatorship, engaging directly with communities, and remembering that we are part of a society we can actively shape.

Our Valdosta experience offers a replicable framework: expand competitive pools, provide information to businesses, and build genuine market relationships. The essence of rebuilding American industrial resilience lies in helping every American who builds (atoms or pixels) feel that they are part of the arsenal of democracy. That sense of shared stakes: that your work matters, that your community matters, that you are part of something worth defending, is the foundation of national mobilization. When local businesses gain federal contracts, when manufacturers supply military operations, when communities prosper through defense partnerships, we develop the sense of agency and mutual responsibility that sustains communities and strategic advantage.

Start where you stand. Problems won’t solve themselves while we wait for someone else to act. The cumulative effect of countless individual decisions will determine whether America builds the distributed manufacturing capacity that competition demands. In an era where efficiency without resilience creates vulnerability, the acquisition workforce holds the power to rebuild American industrial strength one contract, one relationship, and one conscious decision at a time.

The views expressed are those of the authors and do not necessarily reflect the official policy or position of the Air Force, Department of War, the U.S. Government, or of any other entities.

Evan Loomis, Jordan Blashek, and Morgan Hitzig are General Partners at Overmatch Ventures.

When founders and national security officials gathered in Washington last week for the Hill and Valley Forum, they shared a common premise: America is in a technology race it cannot afford to lose. They are right. What remains unsettled is how to win it.

The scale of the challenge is worth stating plainly. In 2007, the United States led China in 61 of 64 critical technologies for economic and military power, according to an analysis by the Australian Strategic Policy Institute.

Today, China leads in 57.

For three decades, Beijing has executed a coordinated national strategy to dominate the industries that will define the next century: semiconductors, artificial intelligence, autonomous systems, nuclear energy, quantum computing, and rare earth processing. While Washington debated and Silicon Valley chased the social web, China built a vertically integrated industrial base that is now outpacing America in the sectors that matter most.

The United States still fields the world’s most powerful military. But military power is downstream of technological power. And we are now running from behind.

Defense strategists have a word for how countries respond when a rival pulls ahead in capabilities. They call it an “offset” strategy: a way to gain an asymmetric edge rather than matching an adversary strength for strength.

America has done it before. After World War II, the Soviet Union held vast superiority in troops and equipment across Europe. Rather than match Moscow soldier for soldier, tank for tank, President Eisenhower turned to America’s nuclear arsenal, raising the cost of any conventional attack to an unacceptable level. When the Soviets built their own stockpile, the U.S. offset again in the 1970s by pioneering precision-guided munitions they couldn’t match.

The competition with China is different. Not in its structure, but in its substrate. Great power rivalry is not new. What is new is the degree to which technology has become load-bearing in that competition. The Cold War was primarily an ideological, economic, and military fight, with technology as an important enabler. Wars we fight today may well be determined by technology, and the pace at which the underlying stack is evolving is unprecedented. What’s more, the technologies that matter today are too tightly coupled to focus on winning just one. Quantum computing requires advanced semiconductors. Those chips increasingly rely on AI to design them. AI demands far more cheap, reliable energy than the U.S. currently produces.

Technologies no longer enable competition. They are the competition, and they are compounding faster than any single offset strategy can track. That’s why the next offset strategy must look different.

Founders — not technologies — are the answer. A single bet on a single breakthrough won’t suffice. We must place thousands of bets across thousands of founders, unleashing America’s entrepreneurial ecosystem across every critical frontier simultaneously—energy, AI, space, defense, biotech, advanced manufacturing. Founders are the only organizational unit capable of keeping pace with the rate of technological change. Institutions optimize for known problems. Founders are structurally built for unknown ones. Let the best of them surprise us with solutions central planners could never predict.

Call it the Founders Offset.

America does not need to out-China China. It needs to be more American, not less.

China’s model has real strengths. When the Party decides a sector matters, it can mobilize capital, talent, and political attention at a scale few democracies can match. But state-directed innovation has a ceiling. When the government floods a technology area with resources and mandates outcomes, the result is often what China’s own economists call “involution”—an ever-rising effort producing fewer real gains. You can build a thousand battery factories. You cannot command a breakthrough.

America’s system is architected for exactly the opposite. It is decentralized by design, driven by initiative, and purpose-built for the kind of distributed, high-velocity experimentation that the Founders Offset demands.

Consider our capital markets. In 2024, venture capital investment in the United States totaled roughly $210 billion, compared to approximately $35 billion in China. Even more important though is the incentive structure. In China, founders and investors operate under the constant possibility of state interference driven by political versus profit-maximizing motives. Even if China grew its venture capital by 6x to match ours, China’s model would still disincentivize risk-taking, dissuading founders from pursuing the hardest challenges.

In the United States, a founder who swings and misses can raise again. There is, in fact, a premium for second-time founders, even ones that did not have great outcomes the first time. This is one reason why the U.S. has nearly four times as many unicorns as China. Not because American founders are smarter, but because the American system rewards the kind of asymmetric, long-horizon risk-taking that produces genuinely novel breakthroughs.

Those individual incentives for each founder and investor compound into something larger. At its best, the American system creates a flywheel across government, capital, and industry that no adversary can replicate. Government agencies like DARPA fund breakthrough research. Founders turn that research into technology. Venture capital scales it. Industry and government deploy it. The public markets buy it, which then funds the next generation of breakthroughs.

You can see it already in companies like Saronic Technologies, which is redefining naval autonomy, CHAOS Industries, which is revolutionizing sensors, shooters, and the kill chain, and Impulse Space, which is building the propulsion systems that will define how humanity operates beyond Earth’s orbit. That flywheel is turning again today. The question is how fast America chooses to spin it.

Every generation faces a moment when the stakes of building—or not building—become impossible to ignore.

The next generation of world-changing companies in AI, energy, communications, biotech, and advanced manufacturing will either be built in the United States or selected by the Chinese Communist Party. If tomorrow’s critical technology is innovated, designed, and made in China, we will be living in a very different kind of world—and not one that Americans would accept.

So what can we do?

Policymakers should start with these three things:

1) Expand the loan program offices at the Department of Energy, Department of War, and Small Business Administration to finance the industrial buildout at scale.

2) Remove the regulatory and policy barriers that prevent national labs from partnering with startups and industry.

3) Accelerate the acquisition reform efforts already underway, giving insurgent companies clear short-term pathways to programs of record.

Investors can get creative about how America finances its industrial revival, combining every tool in their financial arsenal, from venture to project finance and the public markets, to accelerate progress and pathways to scale for this new generation of companies.

Scientists and engineers sitting inside universities and national labs can ask a harder question: is publishing research the highest-leverage thing I can do, or is building something new?

And founders—the ones who already feel the pull of a problem they cannot ignore—should pursue those quests that can tip the balance in America’s favor. The resources are there. The markets are real. And if successful, they will not only create generational businesses, they will create generational impact as well.

America never wins by planning its way to victory. It wins by unleashing people who refuse to accept the gap between what exists and what is possible, and then building until that gap closes. The Founders Offset is not a government program or a policy agenda. It is a belief that the most powerful force in the global technology race is the American entrepreneur who decides that this is their problem to solve.

The next offset will be recognized—years from now—as the moment a generation of founders decided that America’s problems were their problems. We believe that moment is now.

Stuart Lodge is the CEO of Lodge Systems, a Canadian company building a domestic, software-defined manufacturing stack for flight-ready composite airframes and UAV fuselages.

The Pentagon is built around the Program of Record. Identify one problem, select one solution, lock the design, then scale it for years. That approach exists for good reasons. It creates accountability, supports training and sustainment, and keeps complex programs safe enough to operate.

Autonomy behaves less like a platform and more like a living ecosystem. The advantage comes from speed of adaptation, not from perfecting one design in a conference room and hoping it survives the next countermeasure. Likewise, in software we do not appoint a single point of truth and ban everyone else. We let thousands of teams build, ship, fail, learn, and compete. The ecosystem becomes a meritocracy because reality selects what works.

Defense hardware often does the opposite. We select early, freeze early, and optimize against a stable set of metrics that can look convincing while drifting away from the messy reality of mud, cold, jamming, maintenance, and exhausted operators. The result is a monoculture: one architecture, one supply chain, one brittle bet.

That may be changing as the United States and its allies race to build arsenals of small drones and throw out their old acquisition playbook.

In the United States, the Drone Dominance Program is targeting more than 300,000 small drones over the next several years, with military planners estimating each Corps will need 1,500 to 2,000 per day during active operations. In Canada, the Army’s MINERVA Initiative is building a framework to integrate small uncrewed systems across land, air, and maritime roles, with an explicit focus on iterative development alongside domestic industry.

The demand signal is real on both sides of the border. What neither country has is a manufacturing base that can move hardware from first prototype to fieldable production without forcing teams to start over with new materials, new tooling, and new suppliers. The gap exists because the West’s two default approaches, additive manufacturing and traditional aerospace composites, each solve half the problem. One iterates but cannot scale. The other scales but cannot iterate.

Technologists often speak about going from zero to one. The urgent challenge today is figuring out how to go from prototype to production. Solving this problem means building a manufacturing stack where the same materials and processes carry a design from unit one to unit ten thousand, so that every flight hour at prototype scale still counts at production scale.

The 1-to-100 Trap

The West does not need a million of one drone. We need the capacity to let a hundred teams build a hundred variants at the same time, then prove which ones survive real use. But we are failing at the most important phase of hardware learning: the jump from Unit 1 to Unit 100.

This is the hardware valley of death. If you cannot build production-representative batches, you compensate with models, lab demos, and a handful of prototypes. Then the first honest stress test arrives in operational use and the learning shows up late, when it is most expensive to change.

U.S. operational test guidance is blunt: decisions tied to full-rate production are supposed to be informed by testing with production systems or production-representative test articles. If you cannot afford to build enough representative units to test, you are not doing the kind of validation the system expects. You are guessing, and hoping the guess holds up under fire.

Why the Obvious Manufacturing Answers Miss the Mark

Drone Dominance and MINERVA set the demand signal. Translating it into hardware means hitting hard manufacturing constraints. Per-unit costs for attritable systems need to land around $5,000. The Army’s SkyFoundry initiative is planning facilities that can produce 10,000 systems per month. At those price points and production rates, the airframe cannot depend on materials or processes that were designed for platforms built to last thirty years. It needs to be cheap, fast, structurally sound, and producible on a line that can change geometry between production runs. Neither of the two manufacturing approaches the defense world defaults to can deliver.

Additive manufacturing is a false summit. It works well for getting to the first prototype, and carbon-fiber-reinforced filaments have pushed 3D-printed parts closer to structural relevance in recent years. But the gap between a printed demo and a fieldable airframe is wide, and it gets wider under exactly the conditions that matter: vibration, moisture, temperature cycling, sustained operational stress. A printed part is built layer by layer, and each layer boundary is a potential failure plane. Moisture absorption in printed composites runs significantly higher than in compression-molded equivalents because the microstructure is more porous. Dimensional consistency drifts from part to part, with shrinkage rates exceeding two percent on larger components. The physics of the process makes this unavoidable.

The rate problem is just as severe. Even the most ambitious deployable additive systems—containerized print factories designed to operate near the point of need—top out at roughly fifty Group 2 airframes per month. That is a meaningful prototyping capability. Against wartime consumption rates measured in thousands of drones per day, it is a rounding error. You cannot 3D-print your way to a wartime production base.

The traditional aerospace approach fails in the opposite direction. The composites infrastructure the West has spent decades building was designed for platforms that fly for thirty years: F-35s, 787s, Global Hawks. That entire stack is optimized for durability and low volume. Invar steel molds can take twelve months to develop. Autoclave cure cycles run hours per part. The supply chains depend on a handful of specialized firms in Japan, the United States, and increasingly China. A CSIS analysis published in late 2025 found that carbon fiber production cannot be surged, and that any disruption would ripple across every composite-dependent program in the defense industrial base. Aerospace-grade carbon fiber is the right answer for a fighter jet that needs to last decades. For an attritable drone that costs five thousand dollars and may not come back from its first flight, it is the wrong material, at the wrong price, from the wrong supply chain.

So the defense innovation ecosystem is stuck between two bad options: a prototyping method that cannot scale, and a production method that cannot iterate.

One company we work with is caught in exactly this gap. They are 3D printing medium-sized quadcopter airframes. It works fine for development, but they hit a wall after about five units. Each part needs extensive support structures and post-processing. The per-unit cost stays stubbornly high. The parts are not production-representative. Their only conventional path forward is to re-engineer the entire system for injection molding, spend fifty thousand dollars or more on permanent steel tooling, and lock a geometry before they have had enough flight hours to know if it is right.

Every team trying to enter the defense UAS space hits the same wall. It is happening right now, across dozens of startups, in the path of every program from Minerva to Drone Dominance to Canada’s IDEaS challenges. The West keeps producing impressive demos that stall before they reach operational relevance. The designs are often good. The manufacturing options force a choice between learning and scaling, and the current industrial base does not let you do both at the same time.

Mass Iteration

Real hardening requires feedback at a volume where failure shows up with enough frequency to act upon. One prototype tells you whether the concept works. Five prototypes tell you whether the build is repeatable. Fifty units, in the hands of real operators, in bad weather, under jamming, is where you actually start learning how a system fails. Below that threshold, you are still guessing.

This is a different manufacturing requirement than rapid prototyping, and it is a different requirement than mass production. Rapid prototyping optimizes for speed to first article. Mass production optimizes for cost at volume. Neither one is designed for the thing that actually matters in a fast-moving conflict: the ability to build, test, break, learn, and rebuild in operationally relevant batches, on a cycle measured in weeks rather than years.

Call it mass iteration. The side that can run this loop fastest will have a compounding advantage. Every batch of fifty reveals something the previous batch could not. Multiply that across dozens of teams running parallel experiments and you get an innovation rate that no single program of record can match, no matter how well funded.

We already know what happens when one side has this capacity and the other does not. Ukraine’s Ministry of Defense reported delivering over 1.2 million drones of various types in the first eleven months of 2024. The designs are evolving month to month. The teams that cannot iterate fast enough get replaced by teams that can. When consumption is measured in mass, the advantage belongs to whoever replenishes and improves fastest.

The same logic applies outside of conflict. Wildfire response, infrastructure inspection, offshore energy, logistics. Any domain where you need rugged, mission-specific airframes at moderate volumes benefits from a manufacturing approach built around iteration speed.

The Real Requirement: No Second Switch

Most hardware teams experience not one valley of death, but two. They prototype in whatever process is fast and available, usually 3D printing or hand layup. Then, if they get traction, they are forced into a completely different manufacturing process to reach production scale. New materials, new tooling, new suppliers, new qualification, new failure modes. That second switch is where startups go to die.

The problem goes deeper than cost or schedule. Every material system behaves differently under load, under heat, under moisture. A part validated through hundreds of flight hours in printed nylon tells you almost nothing about how the same geometry will perform when injection-molded in glass-filled polypropylene. Beyond delaying production, the second switch resets your engineering knowledge to near zero. All the test data, all the failure analysis, all the hard-won confidence in your design goes out the window the moment you change the material.

We see this pattern playing out across the industry right now. Teams printing airframes that work well enough to demonstrate capability, then stalling because their only path to volume requires them to start over with a different process. Some spend months re-qualifying. Others burn through their runway trying to bridge the gap. The ones that make it to production typically do so by raising enough capital to brute-force the transition. The ones that do not have the capital simply stay small, no matter how good their design is.

The goal should be a single manufacturing stack that carries you from Unit 1 to Unit 100 to Unit 10,000 without changing the material system or the process logic. Same material. Same forming method. Same failure modes. Every hour of flight testing at prototype scale becomes data that is still valid at production scale. No second switch. No reset.

I started my company to build this stack for composite UAV airframes. The core idea is to treat tooling the way software treats code: versioned, disposable, and fast to produce.

Our software models the structural loads an airframe will encounter to define the most efficient reinforcement patterns, producing composite parts with the strength of metals at the weight of plastics. From there, we produce tooling in hours rather than the weeks or months that traditional steel dies require. Our material properties are compatible with additively manufactured tooling modeled on high-throughput automotive stamping, letting us calibrate tooling fidelity to the run size rather than committing to permanent steel dies upfront. If the geometry needs to change, we produce new tooling. There is no six-figure commitment locking us to a shape we are not yet confident in.

The structural material is a thermoplastic biofiber composite sourced from Canadian forestry by-products. The entire supply chain is domestic and sovereign. The result is a drop-in airframe: flight-ready, structurally validated, and made from the same material system at unit five as at unit five thousand.

A team using this process does not re-engineer when they move from prototype to production. The material stays the same. The process stays the same. The failure modes stay the same. The only things that change are tooling fidelity and production rate. And because the tooling is transient, the cost of changing your mind is low enough that you can actually afford to learn from the field before you commit.

What the Ecosystem Needs

The defense industrial base lacks the capacity to build fieldable, production-representative hardware in batches of 50 to 500, learn from real operational use, and then scale the survivors without switching materials, processes, or suppliers.

This is the structural gap that programs like Minerva, Drone Dominance, and IDEaS are running into right now. They can fund the design work. They can fund the procurement. What they cannot do is force a hundred drone startups through a manufacturing transition that the industrial base does not yet support.

Building that missing middle is what I’m focused on. But the need is bigger than one company or one material system. The West needs manufacturing infrastructure that treats iteration as a first-class requirement, treated with the same seriousness as cost and schedule. The teams that can iterate in hardware the way software teams iterate in code will be the ones that produce systems capable of surviving real conflict. The ones that cannot will keep producing demos.

We should stop trying to pick a single winner in advance. We should instead build the manufacturing infrastructure that lets many teams compete, fail fast, and scale the survivors. That requires domestic materials, transient tooling, and production systems designed for learning.

The 1-to-100 gap is a mass iteration problem. And it is solvable.

Today is the release date of our book Mobilize: How to Reboot the American Industrial Base and Stop World War III. This book tells the stories of past American mobilizations that helped our country win the defining struggles of the last century. They are stories about innovation, production, and policy. More fundamentally, they are stories about people: visionary builders, courageous entrepreneurs and bureaucrats, heretical outsiders—above all, patriots.

We wrote Mobilize because the last thirty years have been extremely destructive to American defense capabilities. Our adversaries stopped fearing American power, and we lost deterrence. The challenge of how to build the best weapons, fast and in sufficient numbers, was viewed as a niche issue best left to D.C. policy wonks and a dwindling number of legacy defense contractors. Meanwhile, many leaders from the public and private sector focused on making money and technology in China while America’s industrial base withered.

Now we are fighting Iran. The news is dominated by stories about the rapid depletion of U.S. interceptors and other munitions. We’re in a race to complete our goals before we use up too many of our lethal but difficult-to-produce missiles. China watches from the sidelines knowing every missile used is one less available for the Western Pacific. Other production stats are grim. Our aircraft fleet is half as large as it was during the Cold War, with readiness rates hovering around 50 percent; China built more ships in 2024 than America has built since World War II.

The American industrial base is broken, but it means we have a generational opportunity to rebuild manufacturing for the realities of the 21st century. American prosperity depends on it.

Despite these serious handicaps, the American military continues to operate with exceptional skill. During last year’s Operation Midnight Hammer, stealthy B-2s dropped 30,000-pound bunker buster bombs on underground Iranian nuclear sites. This New Year started with Delta Force operators capturing Venezuela’s Nicolas Maduro in a daring night raid with zero U.S. deaths. Less than two months later, U.S. and Israeli air strikes eliminated Iran’s Supreme Leader Ali Khamenei. And now, as American missiles and bombs rain down on Iranian military targets, F-35 fighters are deployed alongside kamikaze drones costing $35,000.

These feats show that America’s warfighters have the esprit and skill to win. What we need to do is reform the broken culture and broken rules in Washington so they have the tools to finish the job.

Successful mobilization will be defined by optionality and speed. The battlefield extends from the factory to the foxhole and we must be able to pivot production and upgrade hardware at the tempo of AI-enabled warfare. Timelines to improve weapons are now measured in hours and days, not months and years.

Behind the kinetic action, the bureaucratic war is just as hot. Secretary of War Pete Hegseth is igniting a flame front of initiatives. The joint requirements process known as JCIDS is dead, poor-performing programs are on the chopping block, and rampant fraud in programs like 8(a) is being rooted out and eliminated. Then there’s the Revolutionary FAR Overhaul to cut unnecessary regulation, Business Operators for National Defense (BOND) to bring the best of private industry to Pentagon acquisitions, and executive orders enforcing the purchase of commercial technology. The pace of change is dizzying—and refreshing.

The Defense Reformation is underway, but enormous inertia and hostility from the establishment could still thwart the genuinely impressive directives from the top. The fight, truly, is just beginning.

We believe in America. We believe in the primacy of people and that heroes still walk among us. To that end, we hope Mobilize fires up the crazy founders and inspires builders from every corner of the country. We need heretics, whether they have two bars or four stars on their shoulder. To quote legendary fighter pilot John Boyd, we need “people, ideas, hardware—in that order.”

We hope you’ll help us spread the word and order Mobilize.

M-Day was yesterday. We Mobilize today.

Max Streeter is a Deployment Strategist at Palantir Technologies serving U.S. government clients.

In his autobiography, Royal Navy Admiral Percy Scott told of a letter written by a fellow naval officer that breached all protocol. In November 1901, a forty-two-year-old American lieutenant bypassed his superiors and wrote directly to President Theodore Roosevelt. Scott called it “a gross act of insubordination for a junior officer.”

Lieutenant William Sims did not deny the charge. He justified it as a necessary act. He wrote to the president because the United States Navy could not shoot.

That conclusion sat awkwardly beside the country’s recent naval performance. The Spanish-American War had ended swiftly and decisively. But a closer inspection found that the accuracy of American guns hovered around four percent. One after-action observer remarked dryly that “Spain exerted all her skill to lose the war.”

The State of Gunnery

For centuries, naval gunfire had been a battle between gravity and swell. A gun crew stood on a rolling deck, waiting for the fleeting second when ship and target aligned. Fire too early and the shell flew high; too late and it sank into blue water. Gunnery was patience masquerading as precision.

In October 1900, then-Captain Scott of H.M.S. Terrible ended the masquerade. By re-gearing gun elevation mechanisms to allow continuous micro-adjustment, he freed the gunner from waiting for the sea’s permission. Add telescopic sights and extended range, and the effect was transformative: accuracy multiplied, rate of fire surged. The outcome was continuous-aim firing.

Despite these dramatic results, the Admiralty demurred. Scott’s comprehensive and technical report, endorsed by Admiral Edward Seymour and supported by Captain John Jellicoe, received no official Admiralty response. It was allegedly turned down by a junior lieutenant at the gunnery school. The innovation was judged against the cost of redesigning gunnery quarters and retraining all gunners across the Royal Navy.

Across the Hong Kong harbor aboard the USS Monterey, Lieutenant Sims immediately grasped the implications. What set Sims apart wasn’t brilliance or technical acumen, but a perspective that saw technology not as a threat to tradition but as a tool for mission.

The Brilliance of Lieutenant Sims

Sims was a middle-of-the-class graduate of the Naval Academy. He struggled with the entrance examinations and, allegedly, the congressman who wrote his recommendation later claimed he almost regretted doing so. After graduating in 1880, Sims spent 17 years at sea aboard numerous ships. During that time, he was shaped by the bureaucracy around him. It was an education in naval custom rather than innovation.

In 1897, his perspective widened. Posted as a naval attaché to Paris, St. Petersburg, and Madrid, Sims officially collected intelligence on the Spanish Navy. Unofficially, he learned a sobering lesson in how far the U.S. Navy lagged Europe’s great powers.

Those powers were not idle. An industrial arms race was underway: armor shifted from iron to KC steel and effective naval gun ranges expanded from roughly 2,000 yards in the 1880s to nearly 10,000 by the mid-1890s. Accuracy mattered more, not less.

Sims saw in Scott’s continuous-aim firing not a curiosity but a naval necessity. The inconvenience of redesigning ships and retraining crews paled beside a more fundamental truth: a battleship exists to hit what it fires at. Lethality is not ornamental.

Sims installed Scott’s system in his own squadron on the China Station. He took steps to retrain his gun crews and test and measure results. Their accuracy improved markedly. He reported his findings to the Bureau of Ordnance and received no response. His appeal could have ended there, alongside Scott’s. Instead, he appealed to higher authority.

Sims’s letter to President Roosevelt came through desperation. “I have within the past few months submitted to the Navy Department a number of reports on foreign target practice,” he pled. The reports concluded “that our marksmanship is so crushingly inferior to inevitably suffer humiliating defeat at the hands of an equal number of an enemy’s vessels of the same class and displacement.”

Sims had seen the future of long-range naval engagements that would characterize the Great War and sought to give the U.S. a head start. He wrote of how he had modified his own fleet but also highlighted the appalling accuracy of U.S. Navy gunnery. Roosevelt, formerly the Assistant Secretary of the Navy, understood immediately. Sims demonstrated that American ships obtained less than 10 percent accuracy, whereas Scott’s ship, H.M.S. Terrible, achieved 80 to 85 percent.

Roosevelt recalled Sims to Washington and made him Inspector of Target Practice, a position created to prove his claims weren’t fiction.

The outcomes followed swiftly. Sims implemented continuous-aim firing alongside a reformed training regime. Modeled off the Royal Navy, he launched an annual gunnery competition. The winning crews’ methods were written up and circulated across the Navy. Within years, the system increased American accuracy by 100 percent and battery effectiveness by 500 percent. An early American report suggested he had elbowed his technology into frame: “The very first practice under his system convinced the authorities that he was right and that much of the gun gear was all wrong.”

In 1909, Roosevelt summarized the progress of American gunnery and attributed the feat to Sims: “our fighting power is at least five times greater than it was before our training had been improved by Commander Sims’ methods.”

The Iconoclast

The historical record only ever suggests so much about a man’s character. What picture we do get of Sims suggests a career that never settled. In 1918, he wrote to a friend that he had “never liked” the Navy and had “never been comfortable in uniform.” There is a certain breed of officer who loves his service enough to irritate it.

Following the Great War, Sims irritated it some more. He publicly refused his Distinguished Service Medal and rebuked the Secretary of the Navy, Josephus Daniels, for altering promotion standards in ways that made them less meritocratic. His low opinion of Daniels prompted him to write a scathing report, “Certain Naval Lessons of the Great War,” which questioned the Navy’s failure to adequately prepare for war, citing inadequate ship construction among other deficiencies. Some scholars suggest that his parochial pursuit of truth is the reason why he was not promoted from Rear Admiral to Admiral.

Sims spoke truth to subordinates and superiors alike, memorably stating to 400 naval officers:

“It is not only the privilege but the duty of Army and Navy officers to direct letters of constructive criticism to their superior officers, and the officer who chooses to accept personal comfort in place of responsibility for such criticism is not only not worth his pay, but he is not worth the powder to blow himself to hell.”

Sims fought two wars throughout his career: one against foreign navies, another against the peacetime inertia and bureaucratic resistance of his own. The latter proved more exhausting.

This meant overturning assumptions and preconceptions as technology evolved. In 1921, he warned the Naval Academy graduating class against institutional calcification around emerging technology. A year later, after observing a demonstration of aerial bombing tests on naval vessels, Sims pronounced decisively: “the battleship is dead.” The man “who taught the navy how to shoot” had already seen the next generation of carrier-based warfare and wasn’t going to wait.

Continuous-aim firing did not shape naval warfare simply by existing. It shaped conflict because someone refused to let it be ignored. There are great innovators like Admiral Percy Scott who engineer innovative technologies. But an innovative technology is nothing if it languishes in obscurity. We must equally esteem iconoclasts like Sims who refuse to let bureaucracy bury the future.

Every generation confronts its own version of the four percent accuracy problem. Sims’s remedy was clear: focus on the outcome and champion the technology through the institutional machinery.

And write the letter anyway.

Daryl Wieland is Senior Director of Partnerships at Palantir and an Adjunct Professor of Finance at George Mason University.

On the morning of June 6, 1944, General Dwight D. Eisenhower possessed more intelligence data than any military commander in human history. Weather models, tide charts, reconnaissance photographs, intercepted communications, resistance reports, and statistical analyses of German defensive positions flooded his headquarters. Yet when the meteorologists’ models showed a narrow window of acceptable weather, contradicting earlier predictions, Eisenhower faced a decision no algorithm could make for him. He understood something that transcends data: context matters.

OODA Loop

Today’s military planners operate in an environment that would have seemed like science fiction to Eisenhower. America possesses computational power that can simulate thousands of battle scenarios in seconds. Our models evaluate satellite imagery, signals intelligence, open-source data, human intelligence, biometric information, and real-time sensor feeds to generate probability assessments with decimal-point precision. We can model logistics chains, predict maintenance failures, and estimate casualty rates with unprecedented sophistication. These capabilities are transformative in nature but also present a profound danger if we rely on them without context or misunderstand how important “under these conditions” is when using algorithms and models.

Ultimately, it’s about decision advantage and tempo. It’s about tightening the Observe, Orient, Decide, Act (OODA) loop that Air Force Colonel John Boyd made famous. It’s about the team that can make better decisions faster winning. Models and automation can massively accelerate your decision cycle. That’s real advantage. But speed without accuracy means you’re accelerating toward the wrong outcome. It is like option c in this graphic where precision without accuracy doesn’t help:

The goal is rapid, high-quality decisions. That requires tight human-machine teaming where the human interprets the context.

Human – Machine Teaming

The danger lies not in the models themselves, but in our relationships with them. The failure mode we see repeatedly is treating this as a zero-sum game. Either we trust human judgment (slow, inconsistent, doesn’t scale) or we trust the machines (fast, consistent, scalable). That’s a false dichotomy. The right answer is integrated teaming where humans and machines amplify each other’s strengths and compensate for each other’s weaknesses.

A good example of successful human-machine teaming is how PayPal combines automated machine learning models that assign risk scores and flag suspicious transactions with human fraud analysts who manually review uncertain cases, adjust custom filters, and provide feedback that continuously improves the AI models’ accuracy.

This hybrid approach allows machines to process vast numbers of transactions at scale while human experts catch context-specific fraud patterns that algorithms might miss, such as social engineering schemes, and helps the system adapt to evolving fraud tactics through their domain expertise.

We have become a civilization intoxicated with quantification, mistaking precision for accuracy and correlation for causation. When a model produces an output, particularly one rendered in clean graphics with confidence intervals, it carries an aura of objectivity that human judgment lacks. “According to the data” has become the model equivalent of “thus saith the Lord,” an incantation that supposedly settles arguments and justifies decisions. But data, like scripture, requires interpretation. And interpretation requires context.

Under These Conditions

Every model operates within boundaries, explicit or implicit. Interpreting data “under these conditions” means acknowledging conclusions drawn from data are specific to the exact environment, models, and variables present during collection. In other words, context is king.

A logistics model built on assumptions about fuel consumption in temperate climates will fail catastrophically in Arctic conditions. A pattern-of-life analysis trained on peacetime behavior becomes dangerously unreliable when populations are displaced by combat. A terrain analysis model that excels at predicting trafficability in open desert may prove worthless in dense urban environments where a single blocked intersection reshapes an entire operation. These limitations seem obvious when stated plainly, yet they fade from view when we encounter the seductive authority of the output itself.

The problem compounds as modeling becomes democratized. Where sophisticated analysis once required expensive specialists and rare computational resources, today’s junior officer can run complex simulations on laptops. This accessibility is valuable as it decentralizes analytical capability and speeds decision cycles. But it also means that the human capital required to interpret results responsibly must be scaled proportionally. As we distribute power we need to also distribute insights that lead to wisdom.

Decision Dominance

When I talk to defense contractors and customers about deploying advanced analytics, I remind them of the importance of human-machine teaming and to remember the human in the solution. Train your analysts. Build feedback loops. Create curious and skeptical cultures that question the model and the inputs.

This is about building decision support for warfighters operating in complex and contested environments. The technology is incredible. But it’s in service of human judgment, not a replacement for it. Get that relationship right and the result is decision dominance. Get it wrong and the decision making will suffer.

Military history offers sobering examples of this phenomenon. During the Vietnam War, Secretary of Defense Robert McNamara brought unprecedented analytical rigor to military planning. Body counts, sortie rates, weapons-destroyed ratios, and hamlet evaluation surveys generated mountains of data suggesting progress. The models said we were winning. Yet these metrics existed within a context the data couldn’t capture: a political war being fought as a military one, a population whose loyalty couldn’t be quantified, and an enemy whose will the statistics fundamentally misunderstood. The data wasn’t wrong; it simply couldn’t answer the questions that mattered.

Using Common Sense

Consider a more recent example: predictive targeting models that assess strike options. These tools can calculate probable collateral damage, mission success rates, and follow-on effects with remarkable sophistication. They are invaluable assets. But they cannot tell you whether destroying a particular target at a particular moment advances strategic objectives, how the strike will be perceived regionally, or whether the second-order effects align with your theater commander’s intent. They operate within a narrower context than the decision requires.

This is where the irreplaceable value of human judgment emerges. Common sense, which as Voltaire noted is not so common, provides the critical mission ingredient. It’s the experienced planner who recognizes that a model trained on historical data cannot account for an adversary’s recent doctrinal shift. It’s the intelligence analyst who understands that unusual patterns in communications intercepts might reflect religious holidays rather than operational preparation. It’s the logistician who knows that convoy timelines need padding because troops are exhausted, even if the fuel and route analysis say otherwise.

Healthy Skepticism Required

The responsibility that accompanies our analytical power is to cultivate this judgment systematically. It means training our people to interrogate and understand their model assumptions rather than simply run model outputs. It means encouraging healthy skepticism even toward—especially toward—results that confirm our existing beliefs. It means creating organization cultures where questioning the data is considered prudent rather than obstructive.

We must teach our warfighters to ask essential questions: What assumptions underpin this model? What data fed these algorithms? What range of conditions was this tool designed for? What factors exist in my operational environment that the model cannot capture? When the model’s assessment contradicts my observations, which deserves greater weight?

The goal is not to reject analytical tools but rather to employ them properly. Models are force multipliers for human judgment, not replacements for it. They narrow uncertainty, reveal patterns, and free cognitive resources for higher-order thinking. But they remain tools, and tools require skilled operators who understand both their capabilities and limitations.

Ultimately It Was the General’s Decision

General Eisenhower wrote two key messages before D-Day. The first was his inspiring “Order of the Day” to the troops on June 6, 1944, rallying them for the “Great Crusade” with confidence in victory. The second was a secret “failure message” taking sole blame if the landings failed, emphasizing bravery and devotion to duty.

The “Order of the Day” boosted morale, acknowledging the tough fight ahead but highlighting Allied strength, while the failure message underscored his immense responsibility, a testament to his leadership under extreme pressure.

Eisenhower ultimately gave the “go-ahead” and launched the massive D-Day invasion based on the weather model’s narrow window, but he made that choice understanding the broader context: strategic timing, alliance politics, troop morale, and operational alternatives. The data informed his decision; it didn’t make it. That distinction, preserved through decades of technological advancement, remains as critical today as it was on that gray morning in Portsmouth.

The machines can calculate. Only humans can command.

Yoel Margolis is a student at Reichman University in Israel. He was previously a Deployment Strategist intern at Palantir.

When Israel’s Merkava (Chariot) tank was introduced in 1979, experts in Israel and abroad disparaged its radically unconventional design, which broke many commonly held assumptions in the industry. The tank’s chief designer—Major General Israel Tal—was written off as delusional.

Four decades and eight wars later, the Merkava, now in its fourth generation, is regarded as one of the best armored combat vehicles in the world. Israel Tal’s image can be found on the “Wall of Greatest Armor Commanders” at the Patton Museum of Leadership, alongside figures such as Creighton Abrams, Erwin Rommel, and Old Blood and Guts himself.

It is said that necessity is the mother of invention—and the Merkava is no exception.

In the mid 1960s, Israel was in desperate need of modern tanks. The British, recognizing an opportunity for collaboration, agreed to sell the Chieftain, their newest tank, to the IDF. In return, Israel became a secret development partner—providing invaluable knowledge from its recent combat experiences. Israel even helped to sell the Chieftain to Iran—then a discreet ally of Israel on matters of defense.

It seemed like a perfect deal. There was only one problem: the British were bluffing.

Succumbing to pressure from Arab states, the British finally came clean in 1969, after years of relying on Israeli expertise. They had no intention of selling the Chieftain to Israel. Instead, they announced several deals with Arab nations including Jordan—at that time a sworn enemy of the State of Israel.

Facing grim intelligence reports of modern Soviet T62s pouring into the surrounding Arab armies, Israel scrambled for a second option. The United States, reluctant to sell their newest tank, the M60, to Israel, agreed to arrange the sale of refurbished M48 Pattons from the West German army, following an overhaul process conducted by the Italian defense company OTO Melara. This deal also collapsed, partially due to noncooperation from communist Italian union workers influenced by Soviet anti-Israel propaganda.

Driven by his frustration over the Chieftain and Patton debacles, Major General Israel Tal, the head of IDF armored corps, came to a radical conclusion: the time had come for Israel to develop its own domestic tank program.

No stranger to controversy, Tal had earned a reputation during his years of service as blunt and uncompromising—promoting his unique perspectives on armored warfare and its central importance in broader strategic planning. His unconventional views created tension with many powerful members of IDF senior leadership. Tal resigned twice from the military—first in 1969 following a dispute with Chief of Staff Haim Bar Lev over a strategic decision in the Sinai desert. Wooed back three years later, he served as Deputy Chief of Staff and Commander of the Southern Command during the Yom Kippur war until he resigned again in 1974—this time because of an argument with Defense Minister Moshe Dayan.

Despite his resignations, Tal did not disappear from the defense establishment. In 1970, following his first resignation, he founded MANTAK (an abbreviation of Merkava and Armored Vehicles Directorate) as a subunit within the Israeli Ministry of Defense. MANTAK’s one mission was to develop an indigenous tank program that would relieve Israel’s armored corps from any dependence on potentially unreliable foreign states.

Surprisingly, given his discordant relationship with many in the defense establishment, Tal was essentially given free rein. Reporting directly to Dayan, he structured MANTAK with a model of extreme centralization. He employed just 150 engineers and took part in almost every decision—ensuring that his unique design philosophy was implemented across every step of production.

After many years as an armored field commander himself, Tal had developed a perspective on armored warfare tactics that differed from the mainstream. He viewed tanks as the ideal front-line force while most thought of them as support assets that depended on infantry protection. This shift in tactical thinking formed the basis of his guidelines for the Merkava.

Before Merkava, most tank programs promoted a balanced compromise between speed, firepower, and protection of the crew. Tal, drawing on his experience leading tank troops, insisted on a design philosophy that prioritized the protective qualities of the tank above all else. If the tank was to lead the charge in battle, without infantry protection, then ensuring its resilience to enemy fire was crucial. Furthermore, he believed that tank crews with an added sense of protection would perform better, thereby making up for potential losses in mobility and firepower.

In the most notable departure from standard designs, the engine was placed in front of the crew, rather than behind it. A crew hatch was placed in the rear, allowing crews to evacuate an immobilized tank while being protected from enemy fire by the front facing, heavily fortified, engine compartment. Choices such as these added significant weight to the Merkava, severely reducing its horsepower to weight ratio compared to other comparable tanks. It was a tradeoff Tal thought would pay off.

His perspective emerged from personal study of battlefield reports from Israel and around the world. As military strategists Edward Luttwak and Eitan Shamir note, “[The Merkava] is also the only tank designed by tank soldiers based on their own experiences, including the synthesized experience of Tal’s exhaustive ballistics research…” For instance, one report examined 500 damaged tanks and determined that while penetration of the engine compartment only immobilized the tank 2% of the time, a breach of the crew compartment disabled the tank 100% of the time. This report, among others, reinforced Tal’s decision to provide enhanced protection to the crew.

Israel of the 1970s was still very much a product of its socialist founding and its private defense industry was far from the powerhouse it is today. The Merkava’s design and production processes were run from within the Ministry of Defense itself with Israel’s large defense primes (including IMI, Rafael, and others) limited to manufacturing specific components. To anyone familiar with the typical failures of government-run production projects, this arrangement might seem destined for disaster.

Yet Tal succeeded, likely because he reported directly to the defense minister, essentially removing the lengthy chains of command and bureaucratic red tape that typically stifle efficiency and innovation. The freedom he enjoyed, combined with the relatively small size of his team, enabled him to maintain a rapid development timeline—components were designed in weeks (and sometimes hours), immediately tested and refined, then moved to full-scale production.1

This model of rapid feedback loops and compressed timelines proved remarkably effective. Tal delivered combat-ready tanks in 1979—a timeframe similar to tank programs of countries with significantly more developed industrial bases.

The Merkava drew intense criticism upon its introduction. Skeptics both in Israel and abroad questioned Tal’s design decisions—above all, the front-facing engine compartment, which conventional wisdom deemed a fatal compromise.

Unlike many other heretics featured in this series, Tal’s vindication did not take long to arrive. In 1982, the State of Israel invaded southern Lebanon in Operation Peace for the Galilee. The campaign, remembered today as the First Lebanon War, was the Merkava’s combat debut. It silenced the critics almost immediately.

The Merkava spearheaded Israel’s ground invasion and Tal’s controversial design choices proved their merit under fire. Crew survival rates far exceeded those of previous tanks, even when hit. Demand for the Merkava from armored commanders soared as they recognized the added safety and success rates of Merkava crews.

Throughout the war and its immediate aftermath, Tal and his team at MANTAK gathered field reports and combat data, documenting every aspect of the tank’s performance. When the Merkava Mark II rolled out in 1983, barely a year into the conflict, it incorporated dozens of refinements drawn directly from frontline experience. The rapid iteration that had defined the tank’s development continued through its evolution.

Over the next two decades, the Merkava became synonymous with the IDF, ultimately becoming its sole tank in 2004. Tal retired as special armor advisor to the minister of defense in 1989 but remained unofficially involved in developing all four Merkava variants until his death in 2010.

In the two years since the October 7th Hamas attacks, Tal’s philosophy has been vindicated once again. For decades beforehand, Israeli defense leaders, including multiple chiefs of the general staff and defense ministers, had backed strategic policies that marginalized the armored corps. Armored battalions were cut as resources shifted to the “sexier” elite special forces and cyber units. When hostilities erupted in Gaza and Lebanon, the armored corps returned to center stage, proving crucial to the IDF’s counterguerilla strategy in both invasions. Once again, Tal’s theory of armored warfare’s centrality proved correct in the face of skepticism.

Beyond the enduring relevance of tanks, the story of Israel Tal and the Merkava offers a number of lessons for modern military innovators in Israel and abroad. Western nations face a defense industrial challenge similar to Israel’s in the 1970s: how to rapidly rebuild and reshore the production of critical military capabilities. The Merkava program demonstrates how military bureaucracies can break from peacetime inertia and produce genuinely innovative technologies when national security demands it.

Not all features of the Merkava’s story should be replicated. Establishing a MANTAK-esque government-run production line for defense technology would be a recipe for disaster in most large Western countries. However, there is undoubtedly much to learn from how Tal leveraged this potential nightmare of bureaucratic inefficiency. Tal used his organization’s placement within the Ministry of Defense to gain frictionless access to frontline troops for constant testing and feedback. This rapid iteration cycle, combined with his own combat experience, ensured the Merkava’s design reflected actual battlefield needs—not the theoretical preferences of executives disconnected from war.

Most importantly, the Merkava succeeded because Israel’s defense establishment recognized brilliance even in its prickliest form. Tal deserves immense credit, but so do the IDF and Ministry of Defense leaders who backed him—and who had personally clashed with him in the past. They set egos aside, recognized his unique strengths, and empowered him despite the friction.

The question facing today’s defense bureaucracies is whether they can do the same: identify the righteous heretics in their ranks and empower them in spite of—or perhaps because of—their heresy.

Further Reading

The Art of Military Innovation: Lessons from the Israel Defense Forces by Edward N. Luttwak and Eitan Shamir

Nick Miller led Arlington National Cemetery's digital transformation as CIO from 2010 to 2013 following Army service in Iraq and Afghanistan. He then spent six years at In-Q-Tel and six years at AWS Marketplace helping healthcare, non-profits and government organizations access commercial technology.

In 2005 I celebrated Christmas and New Year’s at one of Saddam Hussein’s palaces in Iraq. The generation that went to war after September 11th is in our 30s, 40s, 50s now. We carry the weight of injuries, lost friends, and lost youth.

We also know we lost soldiers to an enemy using cell phones while our institutions took years to field basic technology that could have saved many.

In June 2010, I found myself at Arlington National Cemetery, where our nation’s fallen are put to rest. I was 30 years old, only a few months removed from 24 months in Iraq and Afghanistan, having just dropped out of a PhD program bound to teach at West Point. I was scared, concerned, and confused about my future. Now I was watching us bury the best of my generation—and discovering that Arlington Cemetery’s antiquated systems simply weren’t up to the task.

Families discovered that the graves of their loved ones might be mismarked, misplaced, or missing altogether—headstones pulled from the earth and left mud-caked along a stream, paper records saying one name while the headstone said another. All this on our nation’s most sacred ground.

As a veteran, as an American, that reality was both heartbreaking and infuriating. It was a failure of leadership and of systems that had been left in the past. The root cause was painfully simple: a 150-year-old institution conducting 7,000 funerals a year was still largely running on typewriters (yes, really), 3x5 cards, and tattered paper maps. Senator Mark Warner said it best: “We are one fire, one flood from the records being lost.” Congress was considering taking Arlington from the Army.

I was privileged to lead Arlington’s digital transformation to bring modern technology to military families and the cemetery’s more than 4 million annual visitors. We moved operations off millions of pages of paper records, filing cabinets, typewriters, and fax machines. We had to throw out technology that had been retired decades earlier by industry. In the process, I found personal healing by throwing myself into a valuable mission. I also learned four enduring principles that I hope can guide leaders today.

We can and did do better then, when our nation’s fallen and their loved ones needed it most. We can still do better now.

1. Start Necessary Movement. Choose Direction Over Precision.

Army procedures teach every soldier: “Receive the mission, issue a warning order, make a tentative plan, start necessary movement.” Movement matters more than perfect planning.

At Arlington, we didn’t have the luxury of waiting 12-18 months for requirements. Families were calling every day. 7,000 funerals a year could not wait. We started by understanding real pain—parents wanting assurances their children were in the right place, families unable to find relatives, visitors lost across 624 acres. Then we moved quickly with strategic intent:

• We prioritized the visitor experience by launching a mobile app, ANC Explorer, so visitors could use the phones already in their pockets to navigate and honor their loved ones, even when the enterprise’s standard was still BlackBerry.

• We used available commercial technology, accepting its limits, instead of spending years building aspirational custom geospatial infrastructure specific to cemetery operations.

• We focused on clean data through digitized workflows and distributed stewardship across the organization instead of delaying digital operations while a centralized team created a pristine database from paper.

The lesson: in high-stakes missions, movement with clear and strategic intent beats elegant plans that never reach the people who need them.

2. Executive Leaders Must Drive Speed Through Personal Ownership

On my first day, I went to the Secretary of the Army’s office and met with his executive leadership team—an Army captain in a room with the equivalent of the Army’s senior executives. The Deputy CIO made it clear the mission mattered more than bureaucracy: “Both feet on the gas. Keep it between the white lines. Call ME if you need help.” He backed that up by clearing obstacles himself instead of hiding behind layers.

When some argued to delay launching ANC Explorer until every record was perfectly validated, leadership decided families couldn’t wait—we launched with safeguards and improved from there. Developers worked round the clock with special focus on Section 60—the Iraq and Afghanistan burials—to ensure we got those records right before launch. We listened to feedback and improved in the open.

The result wasn’t a perfect product, but the feedback we received speaks for itself. One caller said simply: “Thank you. Years ago I came to visit Ira Hayes and was unable to locate the grave site. I was finally able to find Ira Hayes.” (World War II veterans and Marines will recognize Ira Hayes from the iconic flag-raising photo on Iwo Jima.)

The lesson: if executives do not personally champion speed and absorb risk, the organization will default to comfort and politics, not mission.

3. Enable Your Innovators: Shared Learning Over Central Planning

On paper, Arlington’s transformation looked “wrong” to traditional governance: no big program office, no five-year architecture roadmap, no fully validated requirements list. On the ground, it looked exactly like other successful technology products: small teams, tight feedback loops with stakeholders and customers, decisions made close to the mission.

• We unlocked innovators at every level. I was 30 years old, fresh from deployment, given the privilege to lead. Every organization has dozens like me; the question is whether your culture enables them to act.

• Everyone understood the vision, clean data requirements, and urgency—enabling rapid decisions across the organization, not just at the top.

• Culture shifted from ‘we can’t’ to ‘yes, here’s how’—systematically removing blockers and enabling action at scale.

The lesson: if you want responsible speed, you cannot centralize every decision. Build shared clarity, then trust your people to move.

Your innovators—your engineers, your product teams, your operators closest to customers—already know what needs to change. The question is: will your processes allow them to act on that knowledge?

4. Embrace Partnerships. Lead With Humility.

When the crisis broke, it would have been easy to hide behind bureaucrat-speak: “the work is inherently governmental—only the government can do this.” Instead, leaders opened the doors. The Northern Virginia Technology Council, industry partners, veterans groups, and technology companies all contributed expertise through a pro-bono assessment and ongoing collaboration.

The partnership model—government mission clarity, industry execution speed, outside perspectives—became a force multiplier beyond what any single team could achieve alone. Not every recommendation was adopted, but every conversation helped refine the path and avoid reinventing wheels others had already built.

The lesson: humility is not weakness. It is recognizing that the mission is bigger than any organization, any process, any individual. That wisdom makes transformation possible. Partnerships make it a reality.

Where Are Your Organization’s Typewriters?

When I tell people Arlington was using typewriters in 2010, they laugh. It’s the natural response to such an absurdity. But today’s organizations are running on their own typewriters. They’re maintained by institutional inertia—the defaults that choose comfort and control over the continuous innovation the age of AI demands:

• Waiting for perfect information instead of moving with clear direction

• Delegating transformation to committees instead of leaders personally owning speed

• Centralizing every decision instead of trusting people closest to customers

• Going it alone instead of listening and building partnerships across sectors

Arlington’s transformation didn’t happen because we chose better. Congress forced it. Then the Army moved decisively, installed new management, and demanded change.

You don’t have to wait for tragedy.

For governments and corporations facing fierce competition, adversary innovation cycles, and missions that can’t afford delay, the stakes are too high to wait. Your systems will break catastrophically or your competitors will win.

Those who internalize the lessons above can guide their organizations to success and victory: Start necessary movement. Leaders personally own speed. Enable your innovators. Build partnerships with humility.

Will your leaders make these choices before they’re forced to?

Our Best Days Are Ahead

Over the next three years, my team retired the typewriter, launched one of the government’s first mobile apps, and helped restore public trust in the management of Arlington National Cemetery. The technology now supports more than 20 other cemeteries, and the Army retained the privilege of administering Arlington.

On occasion, I now use the same technology to virtually view the West Point gravesite of my grandfather, Sergeant Major (Ret.) Thomas Ciurczak, whose funeral I was unable to attend due to my deployment to Afghanistan.

Generations before us rose to their moment. In 1940, FDR called on America to become the Arsenal of Democracy and we retooled an entire industrial base. Today, the generation that deployed after September 11th is rising—alongside entrepreneurs building at AI speed, technologists unlocking commercial innovation, and leaders across sectors who refuse to accept failing institutions.

We’re not waiting for permission. We’re not waiting for perfect plans. We’re moving with strategic intent, unlocking innovators, and building the America worthy of those we laid to rest at Arlington.

Our best days aren’t behind us. They’re ahead. And we’re building them now.

Eliot Pence, Michael Brown, and Jason Nichols are the co-founders of Supply Energetics, a new energetics upstart reshaping the American defense industrial base. In addition to their roles at Supply Energetics, Brown is a General Partner at Bowery Capital, Pence is a founder of Dominion Dynamics and formerly led global growth at Anduril Industries, and Nichols is the former director of AI for Walmart.

Over the last two years, the United States and its NATO allies have committed extraordinary resources to a problem it believed was largely solved: ammunition. The return of high-intensity industrial warfare in Europe triggered an emergency response in Washington, with Congress directing billions of dollars toward what was framed as an urgent need to “surge” munitions production.

Measured narrowly, the effort has produced visible gains. Output of 155mm artillery shells has increased and contracts have been signed (though output has not come close to the desired goal). Yet, today, the broader condition of the munitions enterprise remains fragile. Stockpiles are thin, lead times remain long, and production capacity remains vulnerable to disruption.

Industrial atrophy accumulated over half a century cannot be reversed simply by injecting capital into legacy systems. The United States is attempting to sustain a modern war of mass and precision using an industrial architecture designed for a different era, under different assumptions about scale, safety, and tempo.

The difficulty lies in diagnosis. Ammunition production has been treated primarily as a procurement problem rather than as an industrial systems problem. Funding has flowed, but the underlying structure through which that funding is converted into output has changed little.

The solution involves redesigning industrial production around small, standardized modular units rather than relying on large, centralized high-hazard plants. Using continuous-flow chemistry and automation, these modular systems scale by replication, improve safety and resilience, reduce transport risks, and allow capacity to grow without overstressing existing infrastructure.

Structural Fragility in the Munitions Base

A central weakness of the current approach is the persistence of single-spine dependencies. Despite incremental investments, large portions of U.S. munitions production still hinge on a small number of aging facilities that produce critical energetic components such as nitrocellulose or black powder. These sites function as pacing items for the entire enterprise. Their vulnerability is not theoretical. A single fire, environmental shutdown, or equipment failure can halt national output for months. Increasing throughput at such nodes improves peacetime efficiency, but it does not produce resilience under stress.

Compounding this fragility is the persistence of peacetime contracting logic. Even with emergency funding, industry behavior remains shaped by the expectation that demand will eventually fall back to baseline levels. Firms are understandably reluctant to invest in large amounts of redundant or surge capacity without long-term, bankable signals. The result is a system optimized for just-in-time efficiency rather than sustained conflict.

Modernization efforts, meanwhile, have tended toward incrementalism. Legacy facilities have been instrumented with digital sensors and compliance tooling, but the underlying production processes—many of them conceived in the mid-20th century—remain largely intact. These efforts produce marginal gains, not structural change.

Energetics as the True Bottleneck

The deeper issue is clear when examining the energetics layer of the munitions stack. Energetics—the explosives and propellants encased in shells, missiles, and rockets—are not simply another industrial input. They are a distinct class of material governed by uniquely restrictive handling and transportation standards. Finished explosives and many energetic intermediates are subject to stringent limitations on packaging, routing, shipment size, carrier certification, storage, and transfer.

As a result, energetics do not scale like metals, electronics, or structural components. Even when domestic production exists, movement remains expensive, slow, and capacity-limited, particularly as volumes increase. In crisis conditions, these constraints intensify. Insurance retreats, transport availability narrows, and regulatory waivers struggle to keep pace with operational need. The consequence is that production capacity can become functionally inaccessible long before factories reach physical limits. Throughput collapses not at the point of manufacture, but along the transportation spine that connects industrial nodes.

Reinvent, then Reshore

This dynamic explains why reshoring, by itself, has delivered disappointing results. A domestically produced energetic that must traverse long distances under hazardous-material constraints remains exposed to delay and disruption. The issue is not national origin, but system design. Attempting to move more energetic material through the same constrained transport pathways simply relocates the bottleneck. Without altering how and where energetics are produced, increases in nominal capacity do not translate into usable supply.

Distributed Production as Industrial Redesign

Addressing this challenge requires a shift in industrial philosophy rather than a continuation of current practice. Instead of concentrating output in massive, high-hazard plants, production can be decomposed into smaller, standardized modular units enabled by redesigned synthesis pathways, continuous-flow chemistry, and automation. These systems are safer, more controllable, and scalable by replication rather than expansion.

Modularization distributes capacity geographically, reduces single points of failure, and minimizes transport risk by moving safer inputs—or producing closer to use. The same logic extends beyond energetics: software-defined lines enable rapid reconfiguration, contractor-owned facilities improve operational incentives, and automation increases consistency while reducing human exposure. Collectively, this shifts the munitions enterprise from brittle optimization toward resilience and adaptability.

Aligning Policy and Metrics with Readiness

Policy must follow structure. Episodic appropriations and cost-plus expansion of legacy capacity reward paper output rather than sustained readiness. Performance-based production credits and incentives for validated surge capacity would better align private investment with national security needs, treating readiness itself as a strategic asset.

The core lesson is that industrial design—not just demand—determines wartime responsiveness. Production rate, adaptability, and resilience under disruption now define strategic advantage. The only metric that ultimately matters is reliable production under the conditions conflict actually imposes.

Nathan Mintz is the co-founder and CEO of CX2, an electronic warfare start up. Nathan spent 14 years as an EW and Radar systems engineer at Raytheon and Boeing before becoming founding CEO of directed energy company Epirus and Spartan Radar (recently acquired by John Deere). He has 10 U.S. patents, mostly in RF and EW systems.

“You Need a Lot of Good Yogi Berra Quotes”

The late great Dr. Joe Guerci, former DARPA Special Program Office (SPO) director, friend, and prolific RF guru, once told me he wanted to write a book about “the Art of Straphanging.” For those who don’t know what a straphanger is: it’s a person on a team who is just “along for the ride”:

“It’s a hard job being a straphanger . . . you have to intentionally try to be unproductive . . . you have to love being in meetings and creating useless actions . . . You need a lot of Yogi Berra quotes locked and loaded.” —Dr Joe Guerci

Joe lost his battle with cancer this past year, but I thought I would title this article in his memory.

Joe’s hilarious tirade aside, the fact is that we have too many straphangers in the Defense (now War) Industrial Complex. People getting paid way too much to hang the strap and waste our money, or even worse: to squander it through fraud or abusing the rules to their financial advantage. It’s bankrupting us.

While we should be enthusiastically cracking down on green energy subsidy fraud in North Carolina, Medicaid fraud in Minnesota, and PPP loan fraud in California, we also cannot deny that the Department of War has tens or even hundreds of billions of dollars in waste, fraud, and abuse hidden within. The last report from the DoD Office of Inspector General (OIG) identified over $6.6 billion in criminal fraud and avoidable waste and abuse in a six month period alone.

In this post, I will dive into various schemes of waste, fraud, and abuse I have personally seen in my two decades plus in aerospace and defense that small businesses in particular are guilty of—and talk about what we are already doing or should be doing to rein this in. I could write a book about abuse of cost-plus contracting and earn value accounting, along with the straphanger nature of most SETAs and FFRDCs, so I’ll save those subjects for a later date (maybe my memoirs).

Take a walk with me through a master class in the art of DoW small business straphanging.

Is It Waste, Fraud, or Abuse?

Before we go too far, we need to define our scope and terms. The FAR has very specific definitions of these things, but I’ll start with some pithy definitions that work for me:

• Waste = money spent without necessity or discipline: inefficiency elevated into routine, no malice required.

• Fraud = deliberate falsification for gain: lies told knowingly, paperwork weaponized, intent unmistakable.

• Abuse = rule-following without rule-honoring: compliance stripped of purpose, loopholes mistaken for legitimacy.

It’s important to understand not just what these cheats are, but also the behaviors and glitches in the federal acquisition regulations that cause these to occur.

“If you want to understand the behavior, look at the incentives”

—Charlie Munger

DEI Set Asides

Earlier this month, Secretary Hegseth called out and started a crackdown on one of the oldest examples of this fraud, the 8(a) program for small disadvantaged businesses (SDBs). Here’s a post of his announcement:

Hegseth announced he was taking a “sledgehammer” to the Pentagon’s use of the 8(a) program, and the metaphor is doing real work. The 8(a) Business Development Program—a 48-year-old SBA mechanism built to steer federal contracts toward firms classified as socially and economically disadvantaged—has long occupied an awkward moral niche in Washington: too institutionalized to dismantle, too visibly gamed to defend without qualification. In practice, it became less a ladder than a routing mechanism, a way to move money while preserving the appearance of corrective justice.

In big aerospace, every major proposal came with a small-business volume and a small-business specialist whose job was not to ask whether the structure made sense, but whether it complied. Percentages were allocated, boxes were checked, ownership categories were satisfied—either at the contract level or smoothed out across the corporation.

In too many cases, the small businesses in question are essentially Potemkin contractors: pass-throughs that collect fees while the substantive work is performed by larger incumbents. Many of these firms just exist as toll booths: collecting money that they then dole out in subcontracts to larger incumbents or other firms, but not before collecting their 7-15% fee for collecting the checks and doing the accounting.

I’m seen numerous examples of 8(a) set aside abuse: passthrough contractors with no value add other than providing a contract number and satisfying a small business requirement. The one example of this I’ve seen time and time again is old white guys who put 51% (or more) of the company in their wives’ name to qualify as a woman-owned or woman and minority-owned business and get the contracts directed to them. My own post recounting one experience with it is presently at 1.9 million views and counting on X:

Hegseth’s move does not repeal the law; the statutes remain, the machinery hums on. What it does do is withdraw the Pentagon’s indulgence, forcing large sole-source awards back under scrutiny and making 8(a) a DEI relic of the past.

SBA Administrator Kelly Loeffler sharpened the knife further, cutting required percentages and targeting the cut-out contractors the program quietly trained into existence.

Contract Number as a Service

I’ve had several cases where some DoW customer wanted to buy a particular system from us for a fixed price but was unable to easily because we don’t have a contract number. It’s hard to get a contract number because issuing one is a legal commitment that concentrates authority, money, and personal risk in a single signature, and the system is designed to make delay safer than error. After decades of layered compliance, audits, and scandal-avoidance, contracting has become a ritual of risk deferral where no one owns the whole chain, but everyone can stop it.

So one popular solution is to have a number of “primes” who provide no other service other than to administer a contract number that can be doled out—for a modest fee of something like 6-18% of the topline. The best part is, many of these contracts are also small disadvantaged businesses themselves, so DoW gets credit against their 8(a) requirements to boot. Two birds, one stone!

These sorts of services are a textbook example of what anthropologist David Graeber in his Best Seller Bullshit Jobs refers to as “duct-tapers”: folks who patch over structural failures that should be fixed at the root, such as staff compensating manually for broken software or incoherent processes.

There is room for optimism that the DoW may have found ways around this, however: the Golden Dome program was recently kicked off with a massive Indefinite Quantity/Indefinite Delivery (IDIQ) contract with over 1,000 no-dollar awards. Going forward, the strategy appears to be that individual contractors will simply bid on task orders within this vehicle, cutting out the need for the contract number gatekeepers. This bypasses the need for any of this other stuff.

OTA Consortia Managers: A Different Flavor of Contract Number as a Service

The most popular contract in the new defense era is the “other transaction authority” (OTA), which enables quick bidding and awards while short circuiting the rigorous requirements for “programs of record” that can take years to step through.

Because heaven forbid government contract officers just manage this process itself, the “OTA consortia manager”—basically a non-profit that the bidders have to join and pay fees to—was invented to manage this. The OTA consortia manager oversees the “sources sought” stage which qualifies bidders in the “request for information” phase of major programs. In plain English, we take a layer of bureaucracy out of the DoW and put it in a new entity that shouldn’t exist. Now you have a private sector contractor in a “non-profit” raking in the dough to solve the government’s problem.

The consortia manager doesn’t define requirements, evaluate technical merit, or carry mission risk. It simply talks to the contractors so the government doesn’t have to. That alone places the role squarely in Graeber’s territory for box-ticking and legitimacy theater. Motion without authority, structure without responsibility: a bullshit job not by accident, but by design.

While the job is indeed bullshit, the fees aren’t: OTAs are averaging $10 billion a year in awards lately across the 28 or so consortia. If you assume 50% of awards flow through a consortia with a 1-3% management fee for awards plus management fees, you’re talking hundreds of millions in management fees for these entities every single year.

Service Disabled Owned Veteran Small Business (SDOVSB)

Since we are already slaughtering sacred cows, we might as well go after another one: SDOVSB businesses. While we owe veterans who were injured in the service of our country a debt of gratitude, it’s less clear if using the acquisition rules to pay that debt doesn’t create perverse incentives.

Set-asides and sole-source authorities reward status at the moment of award, not competence over the life of the contract, creating firms whose primary asset is eligibility itself. The service-disabled designation applies at a 0% disability threshold, broadening the category beyond any meaningful proxy for need or limitation. My great uncle Bob Mintz z’’l would show me from time to time the barely crimped range of motion of his right pinky, an injury he earned when he slipped on a ship deck carrying a vacuum tube in the early 50s. That injury gave him “a 0% disability, as declared by a military doctor for which I receive $220 a month, to this day.” A shame he never thought to start an SDOVSB.

The predictable result of this designation is a landscape littered with pass-throughs and cut-outs: nominal ownership satisfying the statute while real work and real risk migrate elsewhere.

Small Business Innovation Research (SBIR) Mills

No discussion of small business abuses is complete without discussing SBIR mills: the bane of Senator Joni Ernst’s existence and a problem that’s currently holding up the SBIR reauthorization act, leaving billions in contracts to legitimate small businesses in limbo.

SBIRs are small contracts from DoW or other agencies with money allocated from the SBA. They are structured in phases, with short, low-dollar Phase I efforts (up to $295k) designed to prove feasibility and larger Phase II awards (up to $1.97M) meant to mature the concept. These contracts provide on-ramps into bigger-money contracts like STRATFI/TACFI for larger prototype and production contracts. Think of SBIRs as non-dilutive pre-seed and seed checks for startups that the government writes to get promising new tech off the ground.

The top 20 “SBIR mills” have consumed $3.4 billion in Phase I and II contracts while often producing little more than elaborate research reports.

—Tyler Beaver, “Defense Companies Get Played on Contracts”

The reality in recent years has been that a small number of contractors have gamed the SBIR proposal process and played a numbers game to capture an excess share of the grants purely for their own profit. This would be fine if they were actually producing viable technologies with that money, but the transition rate is quite poor (17% on average just to Phase II and less than 1% of revenue in some cases coming from later stage Phase III, TACFI, or STRATFI contracts).

According to Ben Van Roo, the top 20 SBIR mills have captured $7.4 billion in defense funding. SBIR mills like Physical Sciences, Physical Optics Corporation (bought by Mercury Systems in 2020), Charles River Analytics and Triton Systems have each accumulated hundreds of awards totaling over $200 million each. These SBIR mills also have an abysmal rate of transition to other contracts (sometimes as low as 1%).

Having hired engineers from and worked with some of these companies before, I’ve seen how they’ve got the grant writing process completely dialed in (and now with the advent of generative AI probably even more so). One engineer who worked as a “chief scientist” at one told me that their operation was so dialed in that they could generate a 20-30% award rate on $150,000 Phase I grants on less than $5,000 in production costs. The “real money” of course is in the Phase IIs, which often have a 40-50% win rate and bring in a million dollars or more for slightly higher proposal costs. The problem is that 80% of these contracts fail to convert to even a Phase II award, much less real meaningful products, sucking funds from more worthy products and contracts. Meanwhile these SBIR mills soak up empty calories that never transition to real products that are relevant and useful to the warfighter.

Another thing I learned about was the “shell game” these companies often play so that when the business gets too large the original owners spin out a portion as a divested entity so that they can qualify for new grants and stay below the 500 employee limit. Indeed, when Physical Optics Corporation was sold off to Mercury in 2020 primarily for its processor and data recorder product lines, the owners prepared for the acquisition by spinning out 7 or more smaller SBIR companies like Intellisense that continued with the same SBIR mill business model (and the same original shareholders), but with a different name on the door.

What Can Be Done?

As Shyam’s 18 theses rightfully point out, “Small Business Programs should not be welfare.” Hegseth’s scrutiny of the 8(a) program is an important first step in reining in one particular area of waste and abuse, but other steps need to be taken to ensure money is spent more efficiently and create an incentive structure that improves the speed and effectiveness of procurement for the benefit of the warfighter and taxpayer.

8(a) contractor set asides should be eliminated entirely at some point—perhaps with litigation through the courts on the civil-rights implications of these programs—but before that long process plays out, there are some mitigations that can be done. Audits should be forced to prove that the owner is an actual performer on the contract and not just a cut-out who is married to the real person doing all the work. Primes should be banned from receiving funding on these contracts from their subs to avoid the kickback cycle I mentioned in my X post on the subject. The president could also go a step further and suspend set aside requirements entirely by EO—rendering the whole scheme moot.

Cut-out contractors can be addressed by more rigorous auditing proving that companies are performing a significant portion of the technical or services work themselves rather than outsourcing—perhaps changing the contract statute to require a hard percentage like 30% or more for smaller performers.

“Contract as a service” shops can be rendered superfluous by streamlining the process for contract awards and making more aggressive use of “cattle car” IDIQs with scores of no dollar awardees like they are using for Golden Dome. This would give everyone a charge number the government can apply funds against, eliminating the need for this particular duct-taper bullshit job.

SVOSDB companies should be subject to more rigorous disability requirements (perhaps increasing the disabled criteria to a higher number than 0%) and/or subjecting them to the same cutout contract rules above.

OTA rules have been expanded so much that I question the usefulness of consortia managers at all. Groups like DIU, RCO, JIATF 401, and others seem perfectly capable of running RFIs, sources sought solicitations, and hiring qualified contract officers to award and manage OTA processes without outsourcing this responsible to external “non-profits” that are just taskmasters anyways.

SBIR mills can be limited by both incentivizing SBIR program managers (known at TPOCs) to get higher transition rates through awards and penalties. Limiting the number of awards per entity per year to reasonable number (like 8) and capping lifetime awards at $25 million or less would also limit the efficacy of the “spray and pray” SBIR proposal writing model. You could also limit the number of SBIR grant applications a company is able to submit each quarter.

We also need to seriously question whether, in the era of billion-dollar companies with 10 employees, the 500-employee limit is too high and should be revisited.

Any of these steps would help push the straphangers out of the way of the performers, save the taxpayer billions, and help focus the farm system of the Department of War’s acquisition apparatus on its core mission: to deliver new capabilities to the warfighter as quickly as possible.

.