Nuclear Reactors on the Moon by 2030: What NASA's Infrastructure Bet Teaches Business Leaders | Robert Braxton

The White House has directed NASA, the Pentagon, and the Department of Energy to put nuclear reactors in lunar orbit and on the moon's surface by 2030. The strategic logic behind the decision is a masterclass in infrastructure thinking that every business leader should understand.

On April 15, the White House Office of Science and Technology Policy published a directive that received surprisingly little business coverage outside the space industry: NASA, the Department of Defense, and the Department of Energy are now formally tasked with putting nuclear reactors in lunar orbit by 2028, and on the surface of the moon by 2030. The headline is dramatic. But the strategic logic behind the decision — the *why* — is a template for thinking about infrastructure investment that applies directly to how businesses build for scale. Why solar power isn't enough at scale** Every current space instrument runs on solar power. It works for small, relatively simple missions. But solar power has a fundamental constraint: it's intermittent, requires bulky battery storage for periods without sunlight, and becomes increasingly impractical as power demands grow. Nuclear fission reactors produce continuous, reliable power for years without refuelling. They're not constrained by solar angle, distance from the sun, or lunar night cycles. The directive targets reactors producing at least 20 kilowatts of electricity for three years in orbit and at least five years on the lunar surface — with designs scalable to 100 kilowatts. The OSTP framing is worth quoting directly: "Nuclear power in space will give us the sustained electricity, heating, and propulsion essential to a permanent presence on the moon, Mars, and beyond." The key word is *permanent*. Solar is adequate for temporary missions. Nuclear is the infrastructure of permanence. The geopolitical dimension: this is about China** The directive doesn't hide the competitive context. China is also developing advanced space energy capabilities, with specific lunar ambitions and a stated goal of establishing a permanent presence by 2030. The US is responding not just scientifically but strategically. This is the same dynamic playing out in semiconductors, AI infrastructure, quantum computing, and undersea cable networks. Advanced nations are making decade-scale infrastructure investments to ensure that the next generation of critical systems is built on their terms, with their standards, using their supply chains. For anyone watching the geopolitical technology competition — and anyone building businesses that depend on technology infrastructure should be watching — space nuclear power is another front in the same strategic contest. The modular, scalable requirement** The directive specifies that reactors must be modular and scalable — designed for expansion rather than fixed at a single capacity point. This is the engineering equivalent of a business building systems that can grow without being rebuilt. It's easy to under-appreciate how hard this is and how often it's done badly. The instinct under time and budget pressure is to build for current requirements. The discipline of building for scalability — designing the 20 kWe system so it can be upgraded to 100 kWe without replacement — requires investing in architecture that won't be fully utilised immediately. Every technology project I've worked on in 30 years has faced this same tension. The businesses that consistently make good infrastructure decisions are the ones that answer the question "what do we need this to do in five years?" before they answer "what do we need this to do today." The contractor competition model** The directive uses a competitive contractor model — NASA and the Pentagon developing energy technologies in parallel, with industry competing for contracts. This approach has a proven track record in space, most visibly in how SpaceX, Blue Origin, and others have driven down launch costs through competition. The parallel development is deliberate. You don't put all your technical bets on one approach when the stakes are this high and the technology is this novel. You fund multiple paths and let capability speak. For businesses making significant technology investments — new platforms, AI systems, infrastructure overhauls — this is a discipline worth adopting. Running parallel proofs of concept before committing to a single vendor or architecture is more expensive in the short term and dramatically less expensive when the chosen path turns out to be wrong. What the 2030 deadline actually means** A medium-power reactor in orbit by 2028 and a functional large reactor on the lunar surface by 2030 is an extraordinarily compressed timeline for nuclear technology in a novel environment. For context, nuclear infrastructure projects on Earth typically take a decade or more. This compression is only achievable because the US has been building the technical foundation — in nuclear miniaturisation, space-rated components, and launch capability — for years before this directive formalised the goal. The 2028 deadline isn't a starting point. It's a public commitment built on existing investment. This is the lesson that gets lost in technology announcements: the moment of visible commitment almost never reflects the moment the real work began. The organisations that are able to move fast when a strategic opportunity arrives are the ones that built the capability before they needed it. The parallel in business infrastructure** I spend a significant amount of my time advising businesses on technology infrastructure decisions — when to invest, what to build, how to architect for scale. The NASA nuclear programme, stripped of its lunar context, is a very pure expression of principles that apply at every scale: Intermittent power (solar) is adequate for limited missions. Continuous power (nuclear) is required for permanent capability. Translate: temporary solutions are adequate for experiments. Infrastructure-grade systems are required for businesses that intend to last. Modular and scalable isn't a nice-to-have. At lunar distances, replacing a fixed-capacity system is not practical. Translate: rebuilding core business systems because they weren't designed for growth is the most expensive thing most businesses do. Parallel development reduces single-point-of-failure risk. Translate: single-vendor dependency in critical infrastructure is a strategic vulnerability, not a simplification. Permanent presence requires a different class of investment than temporary missions. Translate: if you're building a business for the long term, your infrastructure decisions need to reflect that ambition from the start. NASA is betting that the investment required to achieve continuous, scalable power in space is worth making — because without it, nothing else they want to achieve is possible. The same logic applies closer to home. Robert Braxton is CEO of Simple Complex Solutions, an Australian IT consulting and AI strategy firm. For a conversation about technology infrastructure planning for your business, visit robertbraxton.com or call +61 417 226 549.*