
Wunderwaffe (German pronunciation: [ˈvʊndɐˌvafə]) is a German word that literally translates to "wonder-weapon" or "miracle weapon" and was a term assigned during World War II by Nazi Germany's propaganda ministry to some revolutionary "superweapons".Cutaway drawing of a V-1 showing fuel cells, warhead and other equipment.
We Keep Funding ‘Shared Infrastructure.’ Nobody Has Defined It.
Published Nov. 18, 2026 • Read time: 7m
A Patriot interceptor costs about $4 million. The Shahed drone it often gets fired at costs $20,000 to $50,000. In the Strait of Hormuz, Iran has pushed the same math against the world’s most powerful navy, using lightly built fast-attack craft, mines, and fishing dhows rigged for minelaying. The U.S. degraded Iran’s conventional navy, and the strait stayed contested anyway. You can win the battle and lose the financial war.
Our defense industrial base still runs on a handful of large primes making Wunderwaffe. A growing startup scene, in autonomous platforms, is starting to buoy it. Excellence in small numbers becomes a weakness the moment a war runs long. Iran is showing it. The Pentagon’s own inspector general found that munitions spending in the war, which began in February, created inventory shortfalls and exposed bottlenecks in resupply that predate the conflict. Officials say we still have enough for current operations amidst a backdrop of yearslong deficits. Enough for now is not a plan.
Washington has noticed. An April 2025 executive order mandated a rewrite of defense acquisition, A June order targeted drone production. In November the Pentagon told buyers to favor commercial products and direct-to-supplier deals.
Planning for China points the same way. A “Hellscape” defense of Taiwan calls for tens of thousands of cheap uncrewed systems, and Taiwan has proposed buying more than 200,000 one-way attack drones as it only manufactures an estimated 8,000 to 10,000 a year. Policy has moved toward cheap and plentiful. Capacity has not.
To find out why these barriers persist, I spent this summer at Business Executives for National Security researching a term policymakers keep invoking to solve them: shared infrastructure. I had never heard it, so I went looking for the definition.
Funding announcements and investor pitches use the term as if everyone agrees on it. Federal law defines it mostly in telecom, for things like cell tower co-location. For defense manufacturing and research, there exists no statutory or regulatory definition nor an industry standard. So, agencies and investors go on judgment, and expensive facilities get labeled shared without anyone checking.
Writing a definition isn’t hard. I concluded that shared infrastructure is a physical or digital asset built, governed, and financed to give sustained, non-discriminatory access to multiple independent organizations working on public missions. Governed is the word that matters.
Here is why. A shared site cuts the capital barrier, so more small firms can get in. More firms means faster iteration, and faster iteration is how the cost per kill comes down while productivity rises. Primes win too, because they get talent and ideas they struggle to grow in-house or lack the risk-tolerance to implement.
The best case is firms whose work feeds each other, like a port, where shipbuilders, aerospace suppliers, and fuel companies make each other more productive. Three things decide whether a site gets there: tenants who need each other, governance that is clear and accountable, and a community that backs the site and supplies it with talent.
The idea is already being tried. The Pentagon has put $75 million behind a munitions campus in Indiana, next to Crane Army Ammunition Activity, on the premise that pooled equipment lets small firms scale while established manufacturers produce faster. The hub has grown to more than 1,100 acres, with tenants ranging from machine tool makers to solid rocket motor developers.
It is the right instinct and a real test of this argument. It also shows why the missing definition becomes dangerous. The private group behind the hub, ACMI, runs affiliates that hold the Pentagon contract, put up the capital, and act as landlord. Without clear rules, accountability dissolves: no one says who decides which firms get in or what happens if the largest tenant leaves. One tenant’s dedicated parcel is slated to reach roughly 700 of the 1,100 acres. Indiana may ultimately succeed. But right now it exposes the core flaw in current policy: the government is putting tens of millions into “shared infrastructure” without a test to confirm it stays shared without simply benefitting the primary beneficiary on the taxpayers dime.
Governance is the one people skip, and the cases suggest it matters most. Start with Albany. NY CREATES, a nonprofit New York and its university system created in 2018 to manage New York’s high-tech assets, owns and runs the NanoTech Complex under an independent board. IBM, Micron, and Tokyo Electron, rivals, and suppliers, share one of the largest nonprofit chip research cleanrooms in the country. State money built it and a federal CHIPS designation built on it. No tenant can lock the others out.
Now look at Mobile, Alabama. A private equity fund, United Submarine Alliance Qualified Opportunity Fund, assembled 355 acres through an opportunity zone vehicle. Navy oversight rests on side agreements in contracts, not how the site is built. An anchor tenant, Austal, builds submarine modules there, and the fund is still hunting for others. The standard ten-year opportunity zone assessment cycle means someone may eventually want out should the boundaries shift, and no tenant gets a vote. Same kind of asset. Same public support. Only one of these sites protects the firms that depend on it.
Governance has to reach past the fence line, too. A project that hires from local schools and earns the trust of local leaders usually has fewer fights over sitting and permits. Anduril’s Arsenal-1 outside Columbus is a single-owner site, but it shows what the alignment looks like: state incentives, a county port authority asking for $70 million in roads and water, and training programs promised from high school up. Protesters showed up anyway, so support has to be earned and kept.
Talent follows the same rules. People go where the schools are good, the work is steady, and they can afford a home. Groton, Connecticut, where Electric Boat builds submarines, needs more than 6,000 new homes, and home values there are up about 53 percent since 2020. A site that can’t house its workforce can’t grow it.
Three objections deserve a straight answer.
First, a single owner moves fast, and Arsenal-1 is proof. But that speed builds one company’s capacity, and shared sites exist to build the base’s.
Second, primes may capture these sites, or incentives may turn them into tax shelters for investors. Mobile shows how that could happen. The fix is governance, not abandoning the model. With neutral governance and open access, incentives build capacity. Without them, they build real estate.
Third, shared sites are one tool among several. Multiyear contracts and surge orders with primes still matter, but they buy more of what primes already make. They do not bring new builders in.
Security is the harder objection. Classified work, export controls, and trade secrets do not sit easily in multi-tenant space, and some sites should never be shared. MIT Lincoln Laboratory works mostly on classified missions behind base security. It is not shared infrastructure nor should not try to be. The model belongs in the unclassified middle of the base: tooling, testing, and production, where many startups stall.
Congress and the Pentagon should do two things.
First, define shared infrastructure in guidance or statute in the next National Defense Authorization Act, so the term means something once money is attached.
Second, make a simple test a condition of federal funding and incentives.
Does a neutral body govern the site?
Can firms other than the anchor tenant get verifiable access?
Does the funding survive if any one investor or customer walks away?
The Indiana pilot is the place to start: apply the test, publish the answers, and write the standard before the next award. A site that fails may still be a fine facility. It just isn’t shared infrastructure, and taxpayers shouldn’t pay as if it were.
We will not close the gap between a $4 million interceptor and a drone that costs tens of thousands by buying more interceptors. We close it by letting more firms build the cheap answer, faster. That starts with deciding what “shared” means and who is accountable for keeping it that way.
Frederick Gale is an undergraduate at American University writing on counterterrorism, U.S. military posture in the Middle East, and the defense industrial base that sustains it.

Operation Epic Fury and the Illusion of Victory
By Stephen D. Cook • RealClearDefense
Previewing The Shenzen Summit: A Critical View
By Stephen Blank • RealClearDefense
How to Win Without Military Victory
By Kevin Eyer • RealClearDefense
We Keep Funding ‘Shared Infrastructure.’ Nobody Has Defined It.
By Frederick Gale • RealClearDefense
Congress Must Reauthorize Section 702 of the Foreign Intelligence Surveillance Act
By Bryce G. Poole • RealClearDefense

Five Takeaways From Navy Nominee’s Rocky Hearing
By Seligman & Weisgerber, WSJ
U.S. Air Force Taps Anduril for Containerized Launcher
By Global Defense News
Time To Say Goodbye to the NMESIS and NSM
By Marine Corps Compass Points
HII and U.S. Navy Break Ground on Shipyard Complex
By George Allison, UKDJ
Rare U.S. Air Force Spy Plane Spotted Over the Black Sea
By D. Malyasov, Defence Blog
