AI in Defense Manufacturing: Building Autonomous Systems at Scale

How the United States Can Turn AI and Autonomous Weapons into Scalable, Field-Ready Defense Capabilities

Artificial intelligence and autonomous systems are rapidly changing how military forces detect threats, interpret battlefield information, and deploy weapons. In a recent CNBC interview, Anduril CEO Brian Schimpf discussed how these technologies will reshape modern warfare.

The shift is about more than replacing a human operator with a machine. Artificial intelligence (AI) can connect sensors, process enormous volumes of data in real time, identify meaningful patterns, and coordinate multiple systems across contested environments. Autonomous aircraft, maritime vehicles, surveillance platforms, and defensive systems can perform missions at a scale that would be difficult to achieve through conventional platforms and human operators alone.

As AI changes the battlefield, it will also transform the factories and supply chains supporting it. That makes AI in defense manufacturing a national security priority. The central question is whether the United States can build, field, improve, and sustain autonomous systems at the required speed and scale.

From Platform-Centric Warfare to Intelligent Mass

For decades, the United States has relied heavily on highly capable, complex platforms produced in relatively small quantities. Major military systems typically take years to move through development and into production, while some complex platform programs can span decades from initial requirements to full-rate production.

Those platforms will remain essential, but future conflicts are likely to require a different kind of force structure alongside them. Large numbers of comparatively affordable autonomous systems can distribute capability across the battlefield, complicate an adversary’s planning, and reduce risks to American service members. The Department of Defense has used the term “affordable mass” to describe low-cost capabilities designed for production and deployment at scale. Anduril has similarly described its model as “intelligent mass”—combining advanced software with large numbers of distributed, networked systems.

AI makes that scale possible. Modern military operations generate more data than human operators can independently review. An AI-enabled system can combine information from radars, cameras, satellites, unmanned vehicles, and other sensors to create a clearer operational picture in real time. Autonomous platforms can then be assigned objectives rather than requiring continuous manual control.

This does not necessarily remove human judgment from warfare. Instead, AI can allow commanders and operators to spend less time processing raw information and more time evaluating options, establishing priorities, and making consequential decisions. Department of Defense policy requires autonomous and semi-autonomous weapon systems to allow commanders and operators to exercise appropriate levels of human judgment over the use of force. It also requires rigorous verification, validation, testing, system safety, cybersecurity, and understandable human-machine interfaces.

The immediate challenge is translating those requirements and emerging technologies into dependable, producible systems.

Autonomy Is Also an Industrial Strategy

An autonomous system is not merely an algorithm. It is a physical product that depends on processors, sensors, batteries, propulsion systems, actuators, communications equipment, precision components, software, and secure data infrastructure. Each system requires raw materials and components that must move through a complex defense supply chain before they can become finished goods ready for military use.

That makes the expansion of autonomous warfare a defense industrial base challenge.

If the United States plans to rely on large fleets of unmanned systems, traditional defense production processes may not be sufficient. Programs cannot depend exclusively on lengthy development cycles, highly customized components, or fragile supply chains that cannot respond when demand increases. Defense manufacturers need production systems capable of building sophisticated products at commercial-like speed while maintaining the quality, reliability, traceability, and security required by the Department of Defense.

Anduril’s investment in Arsenal-1, its planned hyperscale manufacturing facility in Ohio, reflects this changing requirement. The facility is intended to support the mass production of autonomous systems and weapons through common production processes, flexible infrastructure, and software-enabled manufacturing.

The underlying lesson extends beyond one company or facility: product innovation and production innovation must advance together. Effective AI in defense manufacturing requires companies to design products for scalability, establish resilient supplier networks, and build manufacturing readiness into the development process.

Software Speed Must Be Matched by Factory Speed

AI-enabled defense systems may evolve much faster than traditional military hardware. Software can be updated as missions change, new threats emerge, or operational data reveals opportunities for improvement. Rapid software development, however, creates limited strategic value if manufacturers cannot produce the underlying hardware or incorporate changes efficiently.

Manufacturers need closer integration among engineering, production, supply chain, quality, and software teams. They also need factory systems that can accommodate frequent design iterations without creating constant disruption on the production floor or undermining established production schedules.

Supporting this model requires several capabilities:

• Modular product architectures that allow components and software to be upgraded without redesigning an entire system.

• Digitally connected production environments that provide real-time visibility into quality, raw materials, inventory, work in process, supplier performance, and delivery status.

• Resilient domestic and allied supply chains for electronics, motors, batteries, sensors, energetic materials, and other critical components.

• Faster qualification and testing processes that maintain technical rigor while removing unnecessary delays from the development process.

• Flexible production capacity that can shift among products as battlefield requirements and Department of Defense priorities change.

• A workforce trained to operate at the intersection of advanced manufacturing, robotics, software, and national security requirements.

The companies that succeed will not simply be those with the most advanced AI. They will be those that can repeatedly convert technology into reliable, affordable, and fielded capability.

Scale, Cost, and Attrition Matter

Autonomous systems also change the economics of warfare. A force built entirely around exquisite, limited-quantity platforms can become vulnerable when facing large numbers of lower-cost threats. Missiles, drones, and other systems may be consumed quickly during a sustained conflict, placing tremendous pressure on existing stockpiles, production schedules, suppliers, and manufacturing capacity.

The United States therefore needs systems that are capable enough to accomplish the mission, but affordable and producible enough to deploy in meaningful quantities. Achieving that balance requires disciplined design-to-cost decisions, mature supplier relationships, reliable access to raw materials, stable production planning, and early attention to manufacturability.

Manufacturers must also consider how systems will be repaired, replenished, and upgraded over the long term. Mass production cannot be treated as a final step that begins after a product has completed development. Manufacturing, supply chain, sustainment, and cost considerations must be incorporated throughout the development process.

Demand visibility is equally important. Industrial companies are unlikely to invest in facilities, tooling, equipment, suppliers, workforce, and production capacity based solely on the possibility of future orders. The Department of Defense must communicate requirements clearly, aggregate demand where possible, and provide credible long-term production signals that justify private investment.

Without that alignment, prototypes may continue to advance while the industrial base remains unable to manufacture finished goods at the required rate.

AI in Defense Manufacturing Will Shape the Future of Warfare

Artificial intelligence and autonomy will reshape the battlefield, but they will not eliminate the fundamentals of industrial execution. In many ways, they make those fundamentals more important.

Future military advantage will depend on how quickly the United States can move from software development and successful demonstrations to repeatable mass production. It will depend on whether suppliers can deliver critical materials and components, whether factories can increase output without sacrificing quality, and whether autonomous systems can be repaired, updated, and replenished during prolonged operations.

Preparing the defense industrial base for this future requires more than expanding individual factories: it requires an integrated strategy connecting Department of Defense demand, private investment, product design, raw materials, supply chain resilience, production processes, and workforce development.

AI in defense manufacturing will require shorter innovation cycles, more demanding production schedules, stronger supplier coordination, and flexible manufacturing systems that can respond as operational requirements change. That preparation must begin before a crisis creates immediate demand.

The countries that lead in autonomous warfare will not necessarily be those that develop the first impressive prototype; they will be those who build the industrial systems and production capacity required to manufacture intelligent capability at scale.