AI-Driven Autonomy Transforms Future Warfare Economics

Brandon Tseng, co-founder of Shield AI, offers a stark vision of future warfare, one where autonomous drones, powered by sophisticated AI, will fundamentally alter the economics and dynamics of conflict. This conversation reveals the hidden consequences of relying on traditional, high-cost military assets in an era of increasingly sophisticated, low-cost autonomous systems. It’s essential reading for defense strategists, technologists, and investors who need to understand the seismic shift underway in global security and how early adoption of AI-driven autonomy can create a decisive, long-term advantage.

The Unseen Cost of Conventional Deterrence

The prevailing wisdom in defense has long favored massive, expensive platforms--aircraft carriers, fighter jets--as the ultimate symbols of power and deterrence. Brandon Tseng, drawing from his experience as a Navy SEAL and now as a leader in AI-driven defense technology, argues that this model is rapidly becoming obsolete. The immediate, visible strength of these legacy systems masks a critical, downstream vulnerability: their prohibitive cost and the finite number available. In an environment where adversaries can deploy swarms of inexpensive, autonomous drones, the economic calculus of conflict shifts dramatically.

Tseng highlights how the military has historically relied on these high-value assets to project power and deter aggression. This strategy, however, is predicated on the assumption that adversaries will be deterred by the sheer cost and capability of these platforms. The reality, as demonstrated in conflicts like the one in Ukraine, is that the proliferation of cheap, effective drones changes the game. These systems can overwhelm traditional defenses, not through brute force, but through sheer numbers and adaptability, forcing a re-evaluation of what constitutes true military strength.

"You can think of as our fighter jets are, you know, aircraft carriers as those desktops and laptops. You can think of unmanned systems as mobile phones, and they will become ubiquitous just like mobile phones."

-- Brandon Tseng

This analogy is powerful because it frames the shift not as an incremental improvement, but as a fundamental paradigm change. Just as mobile phones democratized computing power and communication, affordable autonomous drones will democratize battlefield capabilities. The consequence for nations clinging to legacy systems is a potential erosion of their deterrent posture. The immediate advantage of a carrier strike group is diminished when faced with an adversary that can deploy thousands of drones, each a fraction of the cost, capable of overwhelming sensors, disrupting operations, or even engaging targets directly. This doesn't mean expensive assets become useless, but their role and scale must be re-evaluated. Tseng suggests a "hybrid force architecture," where a reduced number of high-end platforms are augmented by millions of autonomous systems. This requires a willingness to accept immediate discomfort--divesting from or reducing reliance on costly legacy systems--for a significant long-term advantage in adaptability and cost-effectiveness.

Hivemind: Navigating the Jammed Battlefield

One of the most compelling aspects of Tseng's vision is Shield AI's Hivemind software, an AI pilot designed to enable autonomous operation even when GPS and communication signals are jammed. This is not a theoretical future problem; it is a present reality on battlefields like Ukraine, where electronic warfare is "brutal." Traditional drones, heavily reliant on GPS for navigation and communication for control, become ineffective or even dangerous when these signals are disrupted.

The conventional approach to this problem would involve developing more robust, shielded GPS systems or more resilient communication networks. Tseng explains that Shield AI’s approach bypasses this entirely. Hivemind functions like a self-driving car's system: it uses onboard sensors (cameras, lidar, etc.) to build a real-time map of its environment and localizes itself based on that map, rather than relying on external signals. This creates a system that is inherently resistant to jamming.

"The way that Shield AI solves it is our AI pilot, which is the easiest way to think about an AI pilot is self-driving technology for unmanned systems. What Tesla is doing, where they've got, you know, their self-driving car with a bunch of cameras and it's building a map of the world, that car is localizing itself using the map that it has built versus being wholly reliant on GPS."

-- Brandon Tseng

The downstream effect of this technology is profound. It allows for persistent intelligence gathering and targeting capabilities in contested electronic warfare environments. Tseng notes that Shield AI's V-BAT aircraft, powered by Hivemind, have been operating successfully in Ukraine under heavy jamming conditions, finding targets hundreds of kilometers away. This capability creates a significant competitive advantage. While other drone systems may be rendered useless, Shield AI's platforms can continue to operate, gather intelligence, and execute missions. This demonstrates how investing in a difficult, fundamental technological problem--autonomous navigation independent of GPS--yields a durable, almost insurmountable advantage when that problem becomes widespread. The conventional wisdom of improving existing, vulnerable systems is bypassed by a solution that fundamentally changes the operating parameters.

The Long Road to Defense Tech Adoption

Tseng's journey with Shield AI underscores the immense difficulty of breaking into the defense technology sector, particularly for startups. He recounts being rejected by 30 venture capital firms in Silicon Valley when trying to raise funds in 2015. The prevailing sentiment was that the government was a "horrible customer" and that defense tech was not a viable business. This resistance highlights a systemic inertia within both the venture capital world and the defense establishment itself.

The immediate consequence of this skepticism was a slow start for Shield AI. However, Tseng's persistence, coupled with the undeniable value proposition of their technology (as evidenced by its effectiveness in Ukraine and with the US Coast Guard), has shifted the landscape. He notes that the culture within the military is changing, with a growing openness to "startup culture" and a willingness to question reliance on legacy contractors.

"Everybody knows you don't do defense. You don't build a business in defense tech. That was the, or in the defense field, they're like, 'The government's a horrible customer.'"

-- Brandon Tseng

The "trick," as Tseng puts it, wasn't a secret switch but relentless persistence--"banging your head against a concrete wall." This requires an entrepreneurial mindset that embraces suffering and pain, viewing it not as a deterrent but as a necessary part of the process. The delayed payoff here is immense. While competitors might shy away from the perceived difficulty of defense contracting, those who persevere and deliver undeniable value can establish a dominant position. The current enthusiasm for defense tech, with billions being invested, is a testament to this long-term play. Companies that endured the initial skepticism and built robust, valuable products are now reaping the rewards. This offers a lesson in patience: the most significant advantages are often built through sustained effort in areas others deem too difficult, leading to a market position that is hard for newcomers to penetrate.

Key Action Items

• Embrace the Hybrid Force Model: Begin planning for a force structure that integrates a reduced number of high-cost, high-capability platforms with a significantly larger, more distributed swarm of affordable autonomous systems. (Long-term investment: 18-36 months for strategic planning and initial R&D).
• Develop Resilient Navigation and Control Systems: Invest in AI-driven autonomy that does not solely rely on GPS or continuous communication, enabling operation in contested electronic warfare environments. (Immediate action: Prioritize R&D in sensor fusion and AI-based localization; payoff in 12-24 months).
• Re-evaluate Traditional Deterrence Metrics: Shift focus from the sheer cost of assets to the cost-effectiveness and scalability of autonomous systems in creating a credible deterrent. (Immediate action: Conduct wargames simulating drone swarm vs. legacy asset scenarios).
• Foster Internal "Startup Culture" within Defense Organizations: Encourage program managers and acquisition executives to actively seek and integrate innovative solutions from non-traditional defense tech companies. (Immediate action: Implement pilot programs with clear success metrics for novel technologies).
• Build Scalable Manufacturing for Autonomous Systems: Plan for the mass production of drones and AI pilots to meet the projected demand for ubiquitous autonomous capabilities. (Long-term investment: 2-3 years for facility expansion and process optimization).
• Invest in AI Pilot Software Development Kits (SDKs): Provide tools and frameworks that enable customers to develop their own autonomous capabilities, fostering a broader ecosystem and deeper adoption. (Immediate action: Develop and release an SDK for internal testing and limited customer pilots).
• Prepare for the "Ubiquitous" Nature of Drones: Recognize that autonomous systems will become as common as mobile phones, requiring a fundamental shift in how military operations are conceived and executed. (Immediate action: Initiate training programs for personnel on operating and integrating with autonomous swarms).