As industrialized warfare collides with rapid technological attrition, defense supremacy will not be determined by who builds the best drone, but by who can outproduce, out-iterate, and out-adapt the fastest. Static acquisitions and niche innovation are insufficient. What is needed are Defense Foundries: integrated production ecosystems that merge engineering, automation, and software to manufacture and evolve entire cohorts of uncrewed systems on demand. Unlike traditional factories built for repetition, foundries are designed for adaptability: engineered to retool quickly, integrate new technologies, and sustain output under pressure. This paper introduces the Defense Foundry Model as a strategic framework for national industrial resilience, proposing a shift in focus from platform procurement to agile production capabilities, aligning military needs, government policy, investor incentives, and industrial capacity under one unifying model.
We call these "defense foundries" to evoke specialized facilities that mold raw materials into foundational, adaptable components. This mirrors the required shift in defense, from making static end-products to investing in production ecosystems forging defense products on demand.
Industrial aspects of emerging conflict dynamics
The argument for the need for such adaptive foundries is based on recent discourse on defense innovation from thinkers such as Eric Schmidt, Eveline Buchatskiy, Mike Brown, Vitaliy Goncharuk, Will Roper, General Mike Minihan, and German Army Lieutenant General Alfons Mais, Martin Feldmann and Gene Kesselman.
While perspectives differ on how to achieve dominance in future conflicts, there's broad consensus on what will define them: automated, robotics-led battlefields and a "war of factories" where scaling production becomes paramount. Efficiency, cost-effectiveness and adaptability in manufacturing are required together with scaling production. This dynamic extends to potential great-power clashes where industrial capacity and cheap, streamlined manufacturing will determine who can outproduce adversaries without economic exhaustion.
While there is optimism for Ukraine's defense innovation model, Vitaliy Goncharuk argues for caution. Their approach is characterized by a network of small start-ups and a "rapid feedback loop between operators and engineers," allowing for daily tweaks that outpace traditional militaries. Goncharuk makes the case that this "zoo of solutions" breeds fragmentation, incompatible systems that falter when scaled, plagued by logistics nightmares, supply chain vulnerabilities, and inability to fulfill large orders. Ukraine's ecosystem, while ingenious and with great cost-benefit ratios, has reached its limits in systemic competition. Russia states it has taken a centralized approach with the aim of a "unified and resilient defense technology stack.” While Ukraine's agile approach sparks innovation, it risks being overtaken by a more concentrated and centralized approach.
This need for urgency has been recognized by senior military leadership, as exemplified by Lieutenant General Alfons Mais, Chief of the German Army, who has critiqued the West's lag in adapting to drone warfare and rapid innovation cycles. In 2024 Mais stated, "We missed this development a little bit. I think we failed more or less altogether to draw the right conclusions. And now we need time to close the gap. The innovation cycles are much too fast for our normal peacetime [procurement] procedures.”
Schmidt writes that this "speed of technological adaptation and iteration" serves as a new and critical measure of combat strength. Goncharuk emphasizes the ideal of infrastructure enabling rapid development, testing, and iteration within a single loop, including direct battlefield integration. These insights underscore the necessity for a new model of defense production. We propose adaptive defense foundries to deliver scalable, resilient ecosystems capable of meeting the demands of modern warfare.
Introducing Defense Foundries
Defense Foundries would be adaptive facilities that embody an engineering, software, and automation trifecta, designed without predefined outputs but with the inherent potential to manufacture diverse systems on demand. Seamlessly integrating evolving technologies via standardized architectures, foundry facilities would be oriented towards the development of inexpensive, autonomous, robotic, and adaptable systems.
At its core, this is a simple concept but with ramifications that are nothing short of revolutionary for the entire defense industry: Defense suppliers should give the government not just the latest drone, but the capability to make iterative versions of the drone. In other words: the military should not just buy drones, but get access to drone foundries. For investors this demands a mindset reset. Instead of focusing on the defense platform to be delivered, investments should target foundry-level capabilities. Being able to produce quickly and adaptively must become the deciding factor when assessing defense technology offerings. Stand-alone product orders run the risk of becoming obsolete by the time they get put to use.
Drawing from recent discourse on defense innovation, there are signals that we are moving in this direction. Eveline Buchatskiy of D3 VC predicts the emergence of "Defense as a Service" (DaaS), a subscription-based model where militaries access ongoing streams of sensors, uncrewed systems, and countermeasures, aligning with modern hardware's disposable, updating nature for constant innovation at predictable costs. This echoes General Mike Minihan's advocacy at the Hudson Institute for "effects as a service" to enhance affordability and agility, shifting the DoD's role from industrial provider to customer of diverse solutions.
Former Defense Innovation Unit Director Mike Brown advocates reforming the Defense acquisition process by budgeting for "capabilities rather than requirements," enabling flexible funding to signal demand and facilitate 12-18 month technology refreshes. Together, these ideas suggest defense foundries could operationalize the shift, providing a capabilities-driven service layer where governments subscribe to resilient production hubs that adapt to threats, blending agile budgeting with embedded software and continuous delivery to foster a war of factories advantage.
Deep Element, a Jordanian defense technology firm, serves as a compelling precedent for the defense foundry model by operating as a vertically integrated innovation hub that partners closely with the Jordanian Armed Forces and the Jordan Design and Development Bureau to design, test, and manufacture drones, counter-drones, and electronic warfare technologies. Rather than merely supplying finished products, Deep Element has established facilities like the Middle East's first dedicated UAS test site in 2022. This enables rapid prototyping, battlefield-integrated iterations, and scalable production to address budget constraints through public-private collaboration. This ecosystem approach allows the military to access an adaptive "factory" for ongoing defense needs.
Key Characteristics of the Foundry Model
While the precise form of the defense foundries remains open for exploration, we offer a sketch of defining characteristics for the Defense Foundry model, inviting the innovation community to build upon and refine this concept.
Defense Foundries could be subsidized production centers, embracing digital engineering and certification with rigorous standards for interoperability. They must also leverage advances in defense manufacturing and on-demand delivery. Optimizing the potential dual-use synergies latent in commercial partners would also yield efficiencies and scale advantages. Finally, the model needs to overcome the tyranny of distance via distributed foundries that enable localized manufacturing and maintenance in remote theaters.
Subsidized production centers: Defense Foundries could be subsidized production centers that are agnostic as to who uses them. Facilities replete with additive manufacturing machines - CNC mills, laser cutters, robotics arms, and advanced simulation software. This model would bring efficiencies into the defense innovation system by giving all relevant companies access to the facilities. These would perhaps operate on a membership or pay-per-use model, open to startups, established defense contractors and partners, fostering a collaborative ecosystem without favoring any single entity.
Another precedent is the DoD's Manufacturing Innovation Institutes (MIIs) that provide subsidized access to state-of-the-art facilities for multiple companies, enabling small defense tech startups to prototype systems or components alongside larger firms, without each needing to invest in their own infrastructure.
Such a model brings efficiencies by lowering barriers to entry, reducing duplication of efforts across industry, and enabling economies of scale in procurement of raw materials or software licenses. In this model, companies compete on design and software, fostering faster iteration and ensuring a steady pipeline of adaptable uncrewed systems
Digital Simulations and Digital Twins: Digital Engineering, i.e. Modeling and Simulation and Digital Twin technologies, must be leveraged. Embracing such advances addresses inefficiencies in defense manufacturing, such as certification processes, over-reliance on physical prototypes and collaboration. Essentially, this is an approach that treats hardware like software, bringing agile into physical systems in their improvement. Drawing inspiration from Formula One racing’s agility, the use of digital twins and simulations allow for virtual testing, certification, and design optimization before physical production. With sufficient tactical environment data points, digital twins can be constructed as precise models of both physical defense systems and their operating contexts. This approach can replicate or enhance the rapid feedback loop advantage between operators and engineers that characterizes Ukraine's defense innovation system. A precedent for this is Will Roper’s ISTARI Digital's collaboration with the USAF on digital twins and digital certification, as seen in their work to develop the world's first digitally certified aircraft using the X-56A platform.
Standards for interoperability: Standards must form a foundational pillar. This approach would ensure interoperability, modularity, and efficient collaboration across diverse producers, enabling the seamless integration of components into adaptable uncrewed systems. Digital Twin technology standards can facilitate digital certification, fostering real-time collaboration to iterate designs rapidly. This aligns with standards like the Modular Open Systems Approach (MOSA), a DoD initiative promoting consensus-based open architectures and the Sensor Open Systems Architecture (SOSA) for sensor interoperability. Producers must agree to common interfaces and data standards to enable plug-and-play integration of hardware and software from multiple sources.
Harnessing Advances in Defense Manufacturing: Defense Foundries must harness cutting-edge advances in defense manufacturing. These are exemplified by companies like Hadrian and Machina Labs. Hadrian deploys AI-driven robotics and automation to create highly autonomous factories characterized by about 10 robots per human worker. These produce aerospace and defense components at roughly four times the throughput of traditional manufacturing models. Hadrian’s Factories-as-a-Service model enables new production hubs to be brought online in under six months. This embodies the foundry's core tenets of agility and scalability. Similarly, Machina Labs utilizes 7-axis robots and AI-driven processes to function like a "robotic vending machine" for on-demand aircraft parts, slashing production times. In foundry applications, this could embed autonomy software at the fabrication stage.
Implementing On-Demand Production: On-demand manufacturing stands as a pivotal requisite for the foundry model, drawing inspiration from the Japanese automobile industry's Just-In-Time (JIT) system pioneered by Toyota. JIT minimizes waste and obsolescence by producing goods precisely when needed through lean, data-driven processes. Rigid batch production would shift to a flexible rolling output model that delivers only the required quantities on a monthly or as-needed basis, while dynamically scaling for the next cycle based on real-time demand forecasts.
Optimizing Dual-Use Capabilities with Commercial Partners: The Defense Foundry model must partner up with the larger commercial sector, to maximize dual-use capabilities latent in automobile manufacturers and big tech firms. Automobile and big tech companies have vast know-how, production scales, and R&D budgets that often dwarf those of traditional defense companies (e.g., Alphabet's $50 billion annual R&D spend compared to Lockheed Martin's $1.5 billion). These capabilities must be harnessed to accelerate adaptive manufacturing ecosystems for scale. Drawing historical inspiration from Arthur Herman's "Freedom's Forge," which details how Ford and General Motors repurposed assembly lines for WWII tanks and aircraft, modern foundries could forge partnerships with auto giants to repurpose facilities for drone production. This approach echoes global precedents, such as Subaru and Kawasaki manufacturing UAVs for Japan's defense forces, Germany's automotive sector pivot to defense amid economic shifts, and French firm Renault collaborating on drone projects in Ukraine.
Big tech firms could bring unparalleled expertise to the Foundry model at the software-hardware nexus. For example, their mastery in AI, cloud computing, and scalable data systems could integrate with physical manufacturing to create autonomous uncrewed ecosystems that evolve through over-the-air updates and real-time analytics.
Overcoming the Tyranny of Distance through Distributed Hubs: The Foundry Model must overcome the "tyranny of distance," the logistical challenges of supplying distant theaters. This could be achieved through a model that includes distributed, forward-deployed manufacturing capabilities that blend containerized hardware, additive manufacturing digital twins, and over-the-air software updates. This would enable localized production, maintenance, and rapid adaptation without relying on lengthy supply chains from distant manufacturing places. For example, deployable "factories-in-a-box" housed in standard shipping containers could be prepositioned at forward operating bases, equipped with 3D printers, CNC machines, and robotics to produce systems, spare parts, or repairs on-demand. Central foundries could maintain digital replicas of deployed assets, ingesting real-time sensor data (e.g., via satellite links) to simulate scenarios, predict failures, and push manufacture adaptions and software patches over-the-air for uncrewed systems' AI, autonomy, or countermeasures.
Conclusion
The Defense Foundry Model shifts defense innovation from product delivery to production agility, an industrial engine that iterates as fast as the fight evolves. Manufacturing must become a strategic asset, shifting from bespoke procurement to scalable, adaptive production ecosystems. This is not industrial policy for its own sake, it is the new terrain of deterrence, where scale, iteration, and adaptability are the arsenal. By aligning these actions, the Foundry Model doesn't just equip militaries for tomorrow's conflicts; it redefines industrial resilience as the ultimate deterrent.
ARTICLE CONTRIBUTED BY
George Howell is VP for Global Industry at RAINCLOUD DEFENSE, accelerating collaboration between defense technology innovators, government stakeholders, and venture capital. Howell operates at the nexus of industrial policy, defense technology innovation, investment and international security. Howell holds an MSc in Development Studies from the University of London.
References
Alphabet Inc. (2024). Research and development spending: approximately $49.3 billion for the calendar year; $52.9 billion for the 12 months ended June 30, 2025 (annualized). Macrotrends. https://www.macrotrends.net/stocks/charts/GOOGL/alphabet/research‑development‑expenses
Bremer, M. K., & Grieco, K. A. (2023, November 8). The Four Tyrannies of Logistical Deterrence. Stimson Center. https://www.stimson.org/2023/the-four-tyrannies-of-logistical-deterrence/
Brown, M. (2025, August 7). Faster capabilities for warfighters with dual defense acquisition. Forbes. https://www.forbes.com/sites/mikebrown/2025/08/07/faster-capabilities-for-warfighters-with-dual-defense-acquisition/
Defense Post. (2024, June 20). Istari secures contract for US Air Force ‘digital twin’ program. The Defense Post. https://thedefensepost.com/2024/06/20/us-digital‑twin‑program‑istari/
Feldmann, M. C., & Keselman, G. (2025, July 17). Factories First: Winning the Drone War Before It Starts. War on the Rocks. https://warontherocks.com/2025/07/factories-first-winning-the-drone-war-before-it-starts/
Goncharuk, V. (2025, August). Ukraine isn’t the model for winning the innovation war. War on the Rocks. https://warontherocks.com/2025/08/ukraine-isnt-the-model-for-winning-the-innovation-war/
Hadrian. (2025, April 18). Hadrian raises $260M to build AI-powered factories for America; adds full product manufacturing, opens Arizona site [Press release]. PR Newswire. https://www.prnewswire.com/news-releases/hadrian-raises-260m-to-build-ai-powered-factories-for-america-adds-full-product-manufacturing-opens-arizona-site-302508179.html
Herman, A. (2012). Freedom’s forge: How American business produced victory in World War II. Random House.
Höller, L. (2025, March 10). Defense companies jack up Germany’s auto industry to make weapons fast. Defense News. https://www.defensenews.com/global/europe/2025/03/10/defense-companies-jack-up-germanys-auto-industry-to-make-weapons-fast/
Hudson Institute. (2025, July 24). Achieving Adaptable Scale: Fielding Military Capabilities as a Service [Video]. YouTube. https://www.youtube.com/watch?v=eGnGBmsna_Q
Institute for Manufacturing, University of Cambridge. (n.d.). JIT Just‑in‑Time manufacturing. Decision Support Tools. https://www.ifm.eng.cam.ac.uk/research/dstools/jit-just-in-time-manufacturing/
Jordan launches first Middle East UAS, CUAS, and EW test site near Kharrana Castle (2022, October 31). Joint‑Forces.com. https://www.joint-forces.com/defence-equipment-news/59227-jordan-launches-first-middle-east-uas-test-site
Lockheed Martin Corporation (2023). Research and development spending: approximately $1.5 billion. TapTwiceDigital. https://taptwicedigital.com/blog/lockheed-martin
Machina Labs. (2025, May 29). Company develops robotic vending machine for aircraft parts. National Defense Magazine. https://www.nationaldefensemagazine.org/articles/2025/5/29/company-develops-robotic-vending-machine-for-aircraft-parts
NSBT Japan. (2025, August 21). Subaru delivers experimental UAV to Japan’s ATLA for manned-unmanned teaming research. Asian Military Review. https://www.asianmilitaryreview.com/2025/08/subaru-delivers-experimental-uav-to-japans-atla-for-manned-unmanned-teaming-research-nsbt/
Office of the Secretary of Defense, Manufacturing Technology Program. (2023, August 7). About the Department of Defense Manufacturing Innovation Institutes (DoD MIIs). U.S. Department of Defense. https://www.dodmantech.mil/About-Us/Manufacturing-Innovation-Institutes-MIIs/
Office of the Under Secretary of Defense for Research and Engineering, Department of Defense. (n.d.). A Modular Open Systems Approach (MOSA). U.S. Department of Defense. https://www.cto.mil/sea/mosa/
OODA Loop. (2025). The future of the Pentagon is digital engineering and Formula One racing-style prowess at this new statecraft. https://oodaloop.com/analysis/archive/the-future-of-the-pentagon-is-digital-engineering-and-formula-one-racing-style-prowess-at-this-new-statecraft /
Le Monde. (2025, June 15). French car maker Renault could make drones in Ukraine, signaling the automotive industry’s shift. Le Monde. https://www.lemonde.fr/en/economy/article/2025/06/15/french-car-maker-renault-could-make-drones-in-ukraine-signaling-the-automotive-industry-s-shift-into-defense_6742366_19.html
Mercury Systems. (n.d.). Sensor Open Systems Architecture (SOSA). https://www.mrcy.com/innovation/capabilities/open-systems-architecture/sensor-open-systems-architecture-sosa
Radin, A. (2025, May). Dawn of automated warfare. Foreign Affairs. https://www.foreignaffairs.com/russia/dawn-automated-warfare
Schmidt, E., & Grant, G. (2025, August 12). The Dawn of Automated Warfare: Artificial Intelligence Will Be the Key to Victory in Ukraine—and Elsewhere. Foreign Affairs. https://www.foreignaffairs.com/russia/dawn-automated-warfare
Sifted. (2025, May 30). What defence in Europe will look like in 10 years — according to 10 experts. https://sifted.eu/articles/defence-europe-10-years-predictions
Times (2024, November 21). NATO’s enemies threaten Western way of life, says German general. The Times. https://www.thetimes.co.uk/article/natos-enemies-threaten-western-way-of-life-says-german-general-d6nh2tz3f
3DPrint.com. (2021, April 12). US Department of Defense contracts ExOne for shipping container 3D printing factories. https://3dprint.com/279178/us-department-of-defense-contracts-exone-for-shipping-container-3d-printing-factories/
