An illustration of the Golden Dome. Credit: Arcfield
By Kevin Kelly,
Published by Space News, 26 Augus 2025
The era of single-vector threats is behind us. That was Aerial Defense 1.0, when defenses were designed for ballistic missiles tracked by radar and defeated by linear kill webs. Threats came one at a time and could be countered in sequence.
We are now entering Aerial Defense 2.0. The battlespace is fast, fused and multidomain. Missiles, jammers, drones and cyber weapons can strike at once from orbit and from the ground. A system built for the last era will not withstand the next.
Golden Dome is more than a missile defense program. It is an opportunity to redefine how we protect the homeland with systems that think faster, scale wider and endure through disruption. But if we cling to legacy assumptions and brittle integration, Golden Dome will fall short before it ever reaches the field.
To meet the moment, government and industry must design from the outset for three imperatives: speed, scale and resiliency.
Speed: compressing the kill web with AI and ML
The days of linear kill webs — detect, decide, respond — are gone. Instead, we are faced with a complex web of sensors, systems and weapons that work together across domains in a coordinated attack. Each handoff, from sensor to operator to weapon, adds latency. And in a new era of hypersonic, coordinated, multifaceted salvos, those delays can be fatal.
The Pentagon’s Chief Digital and AI Office made this urgency clear in its April 2025 guidance, calling for faster artificial intelligence, machine learning and autonomy adoption. The 2025 Defense Science Board interim report echoed this, urging AI/ML integration across the kill web to enable rapid, informed responses.
But policy alone won’t deliver speed. It requires program offices to move AI/ML from slide decks into requirements documents. It demands that AI readiness be measured, not assumed, through real-world testbeds, simulation environments and joint operator feedback loops. Acquisition leads and primes must prioritize AI/ML as foundational to design, not as a downstream integration point.
For Golden Dome, ensuring speed is built in from the outset is a joint responsibility shared across acquisition leaders, program managers and engineering teams. At the leadership level, requirements must explicitly prioritize speed alongside capability and resilience. On the execution side, systems engineers and integrators teams should embed AI/ML-enabled automation, modular architectures and rapid decision-support tools early in the design phase, ensuring human oversight remains in the loop without slowing the operational tempo.
Failing to design for speed means ceding advantage to adversaries who already operate in a faster loop.
Scale: building for a multidomain, multimissile fight
The next conflict won’t unfold in a single domain — it will be a multi-threat, multi-axis synchronized, multi-domain assault. Adversaries will launch blended strikes that combine drones, hypersonics, cyberattacks and electronic jamming, designed to stretch and overwhelm traditional defenses and resource availability, if not allocated judiciously.
The Golden Dome cannot be optimized for just one class of threat. It must be built to integrate space-based sensors, radars, interceptors, cyber defenses and command-and-control frameworks across services and allies, seamlessly and at speed.
There are signs of progress. In June 2025, the U.S. Space Force awarded new contracts through its Resilient Missile Warning and Tracking – Medium Earth Orbit (MEO) program to improve hypersonic missile detection. The new MEO Space Operations Center, opened this summer in Colorado, offering a real-time command hub that connects orbital sensors with terrestrial systems.
But scaling for this fight will take more than better hardware. It requires a digital foundation built for integration and speed. Model-based systems engineering (MBSE), AI-enabled simulations and digital twins must be core enablers, not afterthoughts. These tools let us model full kill webs, stress-test architectures and integrate new technologies faster and with more precision.
They’re not just development accelerators — they’re what turn isolated systems into a unified force. Without them, Golden Dome risks becoming a fragmented patchwork. With them, it can be born integrated, built from the start to scale across domains, programs and partners.
Survival: engineering for disruption and resilience
Modern defense systems must assume one thing: something will go dark. Whether it’s a jammed comms link, spoofed GPS, damaged nodes or cyber breach, resilience means the mission continues anyway.
Resiliency and redundancy begins at the design level, not as a backup plan. That means rerouting data in real time, reassigning tasks dynamically and continuing operations under degraded conditions.
Developing such resilience requires clear ownership. The Department of Defense must set resilience as a hard requirement and define what success looks like. Contractors must carry it through, building modular systems that isolate faults, integrating health monitoring that can adapt on the fly and proving performance through continuous stress tests. Operators close the loop by pushing those designs against real-world conditions. Golden Dome will only work if these roles move together from the beginning.
Defense leaders are starting to put these principles into practice. In 2025, the Pentagon authorized the ADSI C2 gateway to run continuously across tactical networks like Link-16. A $100 million Other Transaction Authority to Anduril is driving development of a tactical data mesh for distributed coordination. And in its FY25 budget, the Pentagon set aside $1.4 billion for CJADC2, whose modular design won praise in a recent GAO review.
These are important steps in the right direction. But true resilience will come only when they are pushed further, stress-tested under contested conditions, scaled seamlessly across domains, and proven to work in real time. Golden Dome gives us the opportunity to accelerate that shift. If we build on this momentum and make resilience a measured requirement, not an assumed one, we can deliver a defense architecture that holds together under pressure and adapts as fast as the threat.
Overcoming structural friction
Golden Dome’s success won’t hinge on a single system or sensor. It will depend on our ability to remove the institutional barriers that keep good ideas from becoming fielded capabilities.
That means rethinking the mechanics of how we deliver national defense.
We need to shift acquisition toward architecture-first outcomes — where performance, integration and speed-to-field matter more than platform ownership. Test and evaluation cycles for AI-enabled and hypersonic systems must be compressed from years to weeks, so capability keeps pace with threat. Data must flow across services and domains without delay, because no system can operate at speed if its operators can’t access the full picture.
Digital engineering tools like MBSE, digital twins and simulation environments are no longer side projects — they’re the backbone of a modern, iterative defense enterprise. We need to fund them accordingly, build them into every program and scale them systematically.
None of these friction points are new. What’s different now is the urgency. Golden Dome gives us a mandate to move faster, connect deeper and field smarter. That will take partnership, leadership and design intent at every level.
See: Original Article
