Defense tech funding is surging, but the real bottleneck isn't AI capability,it's the ability to manufacture weapons at scale. Hadrian, a startup that builds AI-powered factories for defense and aerospace companies, just closed a $1.4 billion funding round at a $7.9 billion valuation, more than quadrupling its worth since January. The company's rapid ascent reveals a critical gap in how the U.S. military is preparing for future conflicts: advanced AI systems are useless if you can't produce the hardware they're supposed to control.

Why Is Manufacturing Suddenly the Hottest Defense Tech Sector?

The Pentagon is facing an unprecedented production crisis. Submarines, munitions, and drone systems are in short supply, and traditional manufacturing methods can't keep pace with demand. Hadrian's Opus software platform, which powers its factories, is already being used by the Army and Navy to automate production workflows and optimize factory operations. The company's CEO explained the urgency: "It enables us to make massive investments to scale our workforce, our software platform Opus that powers our factories, and frankly, get ahead of the massive amounts of production challenges the country has in areas like submarines, munitions, the drone industrial base," Chris Power told CNBC.

explained the

This funding wave is part of a broader defense tech boom fueled by President Donald Trump's plan to reindustrialize the U.S. military and spend an unprecedented $1.5 trillion on defense as geopolitical tensions escalate. But unlike earlier defense tech waves that focused on AI algorithms or autonomous weapons, this one is about the unglamorous work of actually building things at scale.

How Does AI-Powered Manufacturing Change Defense Production?

• Speed: Hadrian's agile development model, demonstrated through a partnership with Palantir, produced an AI command-and-control prototype in just four months instead of the traditional three to five years.
• Integration: South Korean defense firm LIG D&A showcased multiple unmanned surface vessels coordinated through a single AI-based command system, signaling that future exports will bundle AI software with hardware as integrated packages rather than selling individual platforms.
• Continuous Evolution: U.S. defense firms now use subscription-style contracts with the military to continuously update software and models, similar to how Tesla or smartphone manufacturers operate, rather than selling static weapons systems.

The shift toward software-centric defense represents a fundamental change in how weapons systems are developed and deployed. Instead of building a fighter jet or missile system and expecting it to remain relevant for decades, modern defense AI platforms are designed to evolve continuously through software updates. This mirrors commercial tech practices but with far higher stakes.

What Does This Mean for the Pentagon's Command-and-Control Strategy?

The Pentagon is simultaneously pushing to integrate all of these new AI-powered systems into a unified command-and-control architecture called Combined Joint All-Domain Command and Control, or CJADC2. The goal is to create a "network of networks" that securely links all military sensors, data, weapons, and decision-makers across services and combat domains. Palantir's Maven Smart System, which emerged from the Pentagon's now-defunct Project Maven, is becoming the key operational interface for this vision.

To accelerate this integration, the Pentagon recently appointed Army Colonel Molly Solsbury as the Maven Smart System program director within the Chief Digital and Artificial Intelligence Office. Her appointment signals that the Pentagon is treating command-and-control integration as a formal, sustained priority rather than an experimental initiative. Deputy Defense Secretary Steve Feinberg issued a memorandum in March directing the Department of Defense to mature the Maven Smart System into a formal program of record by the end of fiscal year 2026, in September.

The Pentagon's fiscal 2027 budget request included more than $1.5 billion to expand defense users' access to Palantir's Maven Smart System in support of the "Joint Force AI-Enabled Headquarters initiative," which aims to expand enterprise access to joint command-and-control capabilities across all military services and command levels.

Why Are South Korean and U.S. Defense Firms Racing to Build Sovereign AI Platforms?

While Palantir's Maven Smart System is becoming the Pentagon's de facto standard, international defense contractors are simultaneously building alternative AI platforms to avoid dependence on U.S. technology. LIG D&A, South Korea's largest defense contractor, is developing L-NODE, a defense-specific AI platform designed to prevent any leakage of classified defense data. The company is also partnering with LG AI Research to develop command-and-control systems based on large-scale AI models, emphasizing "sovereign AI" architecture that keeps sensitive military information within national borders.

"In South Korea, the mindset is still stuck at buying a car and taking it in for repairs when something breaks," said Jo Gyu-tae, head of LIG D&A's AI Lab. "AI-based weapons systems should evolve the way Tesla vehicles or smartphones do, with software updates that continuously expand their capabilities."

Jo Gyu-tae, Head of AI Lab, LIG Defense & Aerospace

This tension between standardization and sovereignty is reshaping the defense AI landscape. The Pentagon wants unified command-and-control systems that work seamlessly across allied nations, but countries like South Korea are simultaneously building proprietary platforms to protect military secrets and reduce dependence on U.S. technology.

What Happens When Defense Acquisition Can't Keep Pace With AI Innovation?

One of the biggest obstacles to rapid AI integration in defense is the acquisition system itself. Traditional defense procurement assumes that weapons systems are built once and then maintained for decades. But AI systems evolve on a weekly and monthly basis, making it impossible for the defense industry to keep pace using traditional development timelines where core technology typically takes three to five years to develop and full weapons systems can take more than a decade.

LIG D&A's leadership has emphasized that South Korea's defense industry must overhaul its acquisition regulations to support continuous software updates and agile development practices. Discussions on reforming agile processes and acquisition regulations are already underway at South Korea's National AI Strategy Committee and National Assembly. A separate initiative is also in progress to establish five national defense AI transformation hubs that would supply military data to support development.

The broader implication is clear: the countries that can adapt their defense acquisition systems fastest will have the most advanced military AI capabilities. The U.S. is moving faster than most allies, but even the Pentagon is struggling to transition from a hardware-centric culture to a software-centric one.

Hadrian's $8 billion valuation and the Pentagon's $1.5 billion Maven Smart System expansion reflect a strategic recognition that future military dominance depends less on building the smartest AI algorithms and more on building the infrastructure, software platforms, and manufacturing systems that can integrate and deploy those algorithms at scale. The real competition in defense AI isn't happening in research labs; it's happening in factories and command centers.