The new arms race is measured in gigawatts. Modern military AI—from targeting and operational planning to logistics—runs through data centers. Those facilities depend on reliable, round-the-clock electricity supplied largely by private utilities and governed by a fragmented web of federal, state, regional, and local authorities. This means America’s military edge increasingly depends on civilian infrastructure outside of the Pentagon’s control: commercial cloud computing and the electric grid behind it. Our adversaries are taking advantage of this vulnerability. In March, Iranian drones struck Amazon Web Services facilities in the Gulf, disrupting commercial services and demonstrating how military reliance can distribute wartime risk onto civilian companies. An adversary doesn’t need to disable the entire grid to degrade AI-enabled combat power. The threat will scale with the AI buildout. U.S. data-center power use reached 64.4 gigawatts in late 2025 and could approach 183.2 gigawatts by 2030. Meanwhile, more than 2.2 terawatts of proposed generation and storage sit in interconnection queues, where median waits exceed five years. History offers clear warnings about the military significance of energy security. Churchill helped the Royal Navy transition from coal to oil in time for World War I. Rickover removed the endurance constraint on submarines with nuclear power. Germany’s sophisticated wartime industry failed when the fuel system beneath it collapsed. In each case, the decisive question was whether institutions could secure and deliver the most important sources of energy. Today on First Breakfast, Katherine Lambert proposes a defense energy strategy built around site-specific demand forecasts, protected transmission and interconnection rights, federal siting authority for defense-designated loads, and military responsibility for protecting critical commercial cloud infrastructure. The United States has the engineering capacity and capital. The harder question is whether it can build institutions that move at the speed of the threat. Read the full article, “The Arms Race is Measured in Gigawatts,” on our website: firstbreakfast.com/p/the-arm…

Traditional defense procurement systems couldn’t save Ukraine when Russia invaded in February 2022. It built a different one based on rapid iteration, frontline demand, verified performance, and competition. In First Breakfast’s latest interview, UFORCE co-founder and CEO Oleg Rogynskyy explains how Ukraine shifted from buying hardware to buying battlefield outcomes—and how his drone company is scaling Ukraine’s most effective tech and bringing it to the West. Traditional defense programs spend years moving from requirements to production. Ukraine asks users what they need, fields an imperfect system quickly, gathers combat data, and improves it. Rogynskyy argues that future weapons may never reach maturity at the level of an Abrams or F-35: by the time a system approaches “perfection,” the battlefield will have made it obsolete. Ukraine’s acquisition system turns frontline performance into demand signals. Units earn points for verified kills and spend them through the Brave1 Market, where they can compare products, check inventories, read reviews, and request rapid delivery. The government then uses this data to identify successful systems and place larger orders. This is the meritocracy of the battlefield: equipment that works earns its place on the front line. Equipment that fails disappears. Competition produces many options, while operational data helps users and officials separate effective systems from the rest. UFORCE is now trying to bring combat-proven Ukrainian technology—including MAGURA unmanned boats, Nemesis bomber drones, and Liut ground vehicles—to Western markets. After years of stress-testing its products in the most ferocious combat zone on Earth, UFORCE’s new challenge is persuading Western buyers to think more like their Ukrainian counterparts. Will it succeed?

Drones are forcing a fundamental reordering of combined-arms warfare. America’s response must reach beyond production and procurement to the way its military organizes, trains, and fights. Ukraine’s Unmanned Systems Forces was built from the ground up for the new way of war. @DanFMadden and Dom Ventimiglia of @squadra_vc trekked to USF's operations center to see how it works. They met Robert “Magyar” Brovdi, whose unit—the Birds of Magyar—integrate reconnaissance drones, FPV strike drones, electronic warfare, signals intelligence, and battlefield analytics into a single combat system. The results are lethal. Ukraine’s Unmanned Systems Forces achieve 33% of the country’s military’s effects with fewer than 3% of its personnel. In NATO exercises, small Ukrainian unmanned units have obliterated conventional formations, including American armored forces. This lethality is the result of an organizational approach that connects battlefield operations directly to technical talent and industry. Commanders use operations research to assess effectiveness. Engineers iterate capabilities. Verified strikes earn units “ePoints” that they can spend on preferred systems through an Amazon-like procurement platform, creating a feedback loop between battlefield performance and capability development. These arrangements present difficult tradeoffs; they can’t simply be copied by larger forces with global interests, like our own. But they demand urgent study and adoption on our terms. The newly announced AUTOWARCOM shows the United States is moving to develop unmanned forces suited to its own strategic priorities, strengths, and operational challenges. But this is only the beginning of the age of autonomy—a revolution in warfare as far reaching as the longbow and the cannon, and as Ukraine shows more lethal.

Read the full article by @Hunter_morey on our website: The Naval Nuclear Pipeline firstbreakfast.com/p/the-nav…

As America reindustrializes, industries that are starved for technical talent should study the U.S. Navy’s nuclear training pipeline. Under Admiral Hyman Rickover, the Navy proved that ordinary recruits can transform into elite technical operators through rigorous academics, hands-on qualification, relentless standards, and real responsibility. The Navy nuclear model is simple yet demanding. Students first master fundamentals: math, physics, reactor theory, thermodynamics, electrical and mechanical systems, chemistry, and radiological controls. They must explain, calculate, defend, and prove understanding. Then they move to real nuclear plants, where theory becomes action through oral checkouts, evaluated watches, drills, and qualification under pressure. This rigorous training works because of the culture of excellence instilled by Rickover himself. “Good enough” is not good enough. Competence is built through repetition, ownership, hard study, and instructors who know both the equipment and how to develop people. Qualification is only the beginning. Continuous training, audits, daily drills, critiques, and root-cause analysis keep standards from decaying. Civilian industry can’t simply copy the Navy. Companies can’t compel their employees the way the military can. But they can adapt the principles: paid intensive training, labs, simulations, mentored on-the-job qualification, clear standards, frequent evaluation, strong instructors, career ladders, and retention incentives tied to real advancement. Manufacturing, energy, semiconductors, defense production, data centers, and advanced nuclear all need this mindset. If these industries want workers, they will have to grow their own. Rickover’s legacy is about more than nuclear submarines. It is proof that excellence can be engineered. If America wants to rebuild its industrial strength, it must stop waiting for talent to appear and start replicating the pipelines that can produce it.

The White House Office of Science and Technology Policy has released “Science: A New Golden Age,” a comprehensive R&D roadmap that represents the most significant rethink of American science and innovation since Vannevar Bush wrote “Science, the Endless Frontier.” The report challenges how the federal government funds research. And it could shape where American innovation happens. Here’s how. THE PRIMACY OF PEOPLE Instead of channeling money through universities, it proposes directing resources directly to scientists with grants that are portable if recipients move institutions or start companies. It advocates for longer funding horizons and empowers program managers to make substantial bets on breakthrough research, similar to DARPA's portfolio approach. The report argues that the old, institutional model of R&D is bloated and unsuited to developing long-term talent and commercial breakthroughs. Universities extract significant overhead while relying on temporary PhD candidates for core research. Academic incentives prioritize publications over genuine exploration. This model, the report argues, undermines long-term knowledge retention and risk-taking. PROVING GROUNDS FOR INNOVATION Beyond funding, the report emphasizes that innovation requires ecosystems for testing and tinkering. It highlights the importance of testing infrastructure like the proving grounds that developed American aerospace in the last century. New public-private research organizations, mission-focused consortia, and tight integration between scientific breakthroughs and domestic manufacturing capacity are equally essential. PRODUCTION POWER The report also stresses the importance of linking innovation to production. Too often, the United States has invented cutting-edge tech only for foreign countries to scale and produce that tech. American taxpayers ended up funding the prosperity and power of other countries. The lesson is clear: American innovation must translate to American production. The report is sure to provoke the administrators and institutions who depend on the old model of science and innovation funding. But a new era of technological competition calls for new approaches, and this report deserves serious attention from policymakers, researchers, and regional leaders. Read @ThomasLDay’s full assessment of the OSTP report and the opportunity for regional innovation. 👇

Read the full thing—including never-before-posted provocations by our anonymous contributors—on our website: firstbreakfast.com/p/getting…

Chase Dalton: Send a *real* demand signal by putting a price on the wartime outcomes you want.

.@vc: Launch the materials Manhattan Project

Nadia Schadlow: First, check your work. @HudsonInstitute

Will Thibeau: Support the SME industrial base—and let 'em build.

.@ChairmanG: Stop lighting money on fire. Pass the FIRES Act.

Kevin Capozzoli: Accelerate the qualification process

.@ssankar: Open the Magazine

On August 5, Deputy Secretary of War Steve Feinberg lit a fire under industry to solve America's munitions production problem. He sent a memo to industry leaders giving them 21 days to submit plans to “drive significantly faster, more aggressive delivery schedules and/or increased production for critical capabilities. . . " We put the question to some of our favorite doers in the industry: If you had free reign to fix America’s munitions bottleneck, what would you propose? It's been three weeks, so we’re sharing the best—and hottest—takes we received. 👇

The deep sea is an arena of strategic competition. America is behind—but that wasn’t always the case. In the 1970s, U.S. intelligence programs mapped the seabed in remarkable detail. In 1980, Congress passed laws paving the way for deep-sea exploration and mining. The CIA even faked a deep-sea mining expedition as a cover to recover nuclear codes and intelligence from a lost Soviet submarine. But the strategic knowledge faded after the Cold War ended. The statutory authorities gathered dust and were forgotten. China has picked up the baton that America dropped. Beijing has spent a decade quietly locking up the richest untapped resource on Earth: the Clarion-Clipperton Zone, a stretch of Pacific seafloor packed with more cobalt, nickel, and manganese than every known land reserve combined. China’s state-owned COMRA holds 5 of ~30 international seabed exploration licenses—more than any other nation. China has research vessels, deep-sea submersibles, and commercial mining trials already underway. Beijing is racing to unlock the critical-mineral treasure chest hidden beneath the waves. It is comfortably in the lead. The good news: America still has statutes on the books to catch up. The 1980 Deep Seabed Hard Mineral Resources Act lets NOAA issue exploration and mining permits outside the ISA’s slow-moving multilateral process. A 2025 executive order activated it. Crucially, the tech to turn legal authority into reality is maturing fast. American startups are close to commercial-scale proof of concept. This could be deep-sea mining's SpaceX moment, turning existing technology into a new industrial base with focused capital and urgency. The legal authority and tech are here. Will America deploy the capital and demonstrate the political will to explore the deep once again? firstbreakfast.com/p/the-sea…

Analysts and appropriators in the Beltway measure military strength by counting platforms: fighter jets, submarines, aircraft carriers. But a fighter jet without a steady supply of munitions is a liability more than a capability. A submarine stuck in dry dock waiting on spare parts isn’t deterring anyone. As appropriators in Congress consider a historic defense budget, a fundamental mindset shift is needed: Start thinking about manufacturing capacity itself as the strategic asset. Today on First Breakfast, @HudsonInstitute’s Nadia Schadlow and @MarlinspikeVC’s Mislav Tolusic argue that Congress and the Pentagon should adopt a "Capacity-as-a-Service" (CaaS) model to prioritize dynamic production capability over static hardware buys. Borrowing the logic of SaaS, DoW could pay manufacturers ongoing retainers or subscriptions to keep flexible, scalable production capacity on standby—much the same way it already pays commercial shipping companies to guarantee capacity during emergencies. As Congress debates next year's defense budget, the question isn't just what weapons to buy, but what manufacturing capacity to own, fund, or incentivize so industry can produce at whatever scale a future conflict demands. Defense acquisition needs an update to reflect that reality. Read the full argument on our website below 🔗👇