Real problems worth solving

A January 2025 Government Accountability Office report (GAO-25-107283) found that the U.S. defense industrial base has critical dependencies on foreign suppliers for components used in missile systems, and that the Department of Defense lacks adequate visibility into the depth of these dependencies. The problem extends beyond the well-known reliance on Chinese rare earth elements for guidance system magnets and motor components; it includes specialty metals, advanced composites, propellant chemicals, and electronic components sourced from countries that could restrict supply during a conflict or geopolitical crisis. This matters because in a major conflict scenario -- precisely when missile production must surge -- the supply chains for the missiles themselves could be disrupted by the adversary the missiles are meant to deter. If China restricts rare earth exports during a Taiwan contingency, production of precision-guided munitions that depend on rare earth magnets for guidance systems would slow or halt. If European suppliers of specialty propellant chemicals are cut off by a broader conflict, solid rocket motor production faces the same bottleneck. The defense supply chain is only as strong as its most vulnerable link, and the GAO found that DOD does not systematically identify or mitigate these single points of failure. The problem is compounded by the opacity of modern supply chains. A prime contractor like Raytheon or Lockheed Martin may source a component from a U.S.-based Tier 1 supplier, who sources a subcomponent from a Tier 2 supplier in Europe, who sources a raw material from a Tier 3 supplier in China. The prime contractor often has no visibility below Tier 1, and the Department of Defense has even less. When a Tier 3 supplier in a potentially adversarial nation is the sole global source for a critical material, the entire missile production chain rests on a dependency that no one in the procurement system even knows about until it breaks. The structural cause is the globalization of manufacturing over the past three decades, driven by cost optimization rather than supply security. Defense procurement regulations reward the lowest-cost bid, which incentivizes contractors to source globally. The Berry Amendment and other domestic sourcing requirements cover some categories but are riddled with exceptions and do not extend deep into the supply chain. Efforts to reshore critical manufacturing face the reality that building domestic production capacity for specialty materials requires years of investment and a willingness to pay higher costs that the acquisition system penalizes. The GAO recommended that DOD take actions to address risks posed by foreign supplier dependence, but implementing these recommendations requires mapping supply chains to the sub-tier level, investing in domestic alternatives, and potentially accepting higher unit costs for missiles -- changes that conflict with budget pressure and the acquisition system's cost-optimization bias.

The 400 active Minuteman III missile silos spread across Montana, Wyoming, North Dakota, Colorado, and Nebraska were constructed in the early 1960s -- over 60 years ago. The original plan was for the Sentinel program to reuse these silos with upgrades, but the Air Force determined in 2025 that the existing launch facilities are in too poor condition to support a new missile system through the 2070s. The underground cabling connecting silos to launch control centers has degraded beyond the point where it can interface with modern command-and-control systems. The result: entirely new silos and communications infrastructure must be built, which is the primary driver of Sentinel's 81% cost overrun. This matters because these 400 silos and their associated launch control centers are not just holes in the ground -- they are hardened facilities designed to survive nearby nuclear detonations and still function. Rebuilding them means excavating and constructing new hardened structures across five states, running thousands of miles of new hardened communications cable, and doing so while the existing Minuteman III system remains on alert. The Air Force cannot simply shut down the ICBM force while construction proceeds; the deterrent must remain operational throughout the transition, creating an extraordinarily complex construction-while-operating challenge. The workforce problem is acute. Construction of missile silos requires specialized skills in hardened facility construction, electromagnetic pulse protection, and nuclear survivability standards that the commercial construction industry does not possess. The original silos were built by the Army Corps of Engineers during the Cold War when the defense construction workforce was vastly larger. Today, the construction must be performed in remote rural areas of the northern Great Plains, far from major labor markets, in a tight construction labor environment where private-sector projects offer better conditions and pay. The structural reason the infrastructure decayed to this point is that for decades, sustaining silo infrastructure was deprioritized relative to the missiles and warheads inside them. Concrete degrades, rebar corrodes, waterproofing fails, electrical systems oxidize -- but these slow-motion failures are invisible until they reach a tipping point. Annual maintenance budgets kept the silos functional but never addressed the fundamental aging of 1960s-era construction. The assumption that Cold War infrastructure would seamlessly host 21st-century weapons was never rigorously tested until the Sentinel program's engineering review revealed the true condition of the facilities. The consequence is a multi-decade, multi-billion-dollar construction program that must be executed in some of the most remote and harsh environments in the continental United States, using skills the workforce does not currently possess, while maintaining nuclear deterrent operations without interruption. There is no historical precedent for this undertaking.

The United States is simultaneously modernizing every component of its nuclear arsenal for the first time since the Cold War: the W87-1 warhead for the Sentinel ICBM ($15.9 billion), the W80-4 warhead for the Long-Range Standoff Weapon ($13 billion), the W88 Alt 370 life extension, the B61-12 and B61-13 gravity bombs, a new nuclear-armed sea-launched cruise missile, new Columbia-class ballistic missile submarines, new B-21 Raider bombers, and the Sentinel ICBM itself. The Congressional Budget Office estimates the total 30-year cost of nuclear modernization at $1.5-1.7 trillion. Every one of these programs is running concurrently, creating unprecedented demands on NNSA production facilities, the defense industrial base, and the federal budget. This matters because running all programs simultaneously means that any delay or cost overrun in one program cascades across the entire enterprise. The W87-1 warhead cannot be produced without plutonium pits, which NNSA cannot produce at the required rate. The Sentinel ICBM cannot be fielded without the W87-1 warhead. The warhead cannot be tested without facilities that are also needed for the W80-4 program. NNSA's production complex -- Los Alamos, Lawrence Livermore, Sandia, the Kansas City National Security Campus, Pantex, Savannah River -- must execute all of these programs with a workforce that has limited surge capacity and faces competition from the private sector for nuclear engineers and weapons physicists. The fiscal pressure is immense. Nuclear modernization competes for funding with conventional force modernization, cyber capabilities, space systems, and readiness accounts within a defense budget that faces political pressure for constraint. Unlike conventional weapons that can be deferred or canceled, nuclear weapons programs carry existential stakes: if the deterrent fails because a warhead is unreliable or a delivery system is unserviceable, the consequence is not a lost battle but a potential nuclear catastrophe. The structural reason all programs are running simultaneously is that successive administrations since the 1990s deferred modernization to save money in the near term, assuming the existing stockpile would last indefinitely. It did not. Warheads, submarines, bombers, and ICBMs are all reaching end-of-life at roughly the same time because they were all built during the same Cold War production surge of the 1980s. The 'modernization bow wave' was predicted for decades but never addressed until it became unavoidable, creating the current all-at-once crunch. The risk is not that any single program fails but that the aggregate demand on budgets, facilities, workforce, and industrial base exceeds what the system can deliver, resulting in stretched timelines, cost overruns, and gaps in deterrent capability during the transition period.

The cost exchange ratio in modern missile defense has reached unsustainable levels. During Middle East operations in 2024-2025, U.S. and allied forces routinely used SM-2 missiles ($2.2 million each) and SM-6 missiles ($4.3 million each) to shoot down drones costing as little as $20,000-$50,000. PAC-3 MSE interceptors at $3-5 million each were expended against ballistic missiles costing a fraction of that amount. Analysis of UAE defense spending against Iranian-sponsored attacks found that for every $1 Iran spent on drone attacks, the UAE spent $15-35 on air defense -- a ratio that makes sustained defense economically impossible over time. This matters because it inverts the traditional military calculus where the defender has the advantage. If an adversary can force a defending nation to spend $35 for every $1 of offense, they can win through economic attrition without ever achieving a military breakthrough. An adversary need only keep launching cheap drones and missiles in sufficient quantity to exhaust the defender's interceptor stockpile and defense budget. Senior Navy and Defense Department officials described this dynamic as 'unsustainable' by mid-2024, yet U.S. forces continued expending top-tier interceptors against low-end threats throughout 2025 because no affordable alternative was available in theater. The problem cascades beyond immediate cost. Every SM-6 fired at a Houthi drone is one fewer SM-6 available for a high-end contingency against China's anti-ship missile arsenal. Every THAAD interceptor used against an Iranian ballistic missile is one fewer available for North Korean scenarios. The opportunity cost of using premium interceptors against commodity threats depletes readiness for the most dangerous scenarios the U.S. faces. The structural reason this cost asymmetry persists is that the U.S. missile defense architecture was designed during an era when the primary threats were sophisticated ballistic missiles from peer adversaries, not swarms of cheap drones and crude rockets from non-state actors. The entire interceptor development pipeline -- from requirements definition through testing to production -- is optimized for exquisite, expensive, high-performance systems. Programs like directed-energy weapons (lasers) and counter-drone systems promise to close the cost gap, but none are deployed at scale. The Navy's HELIOS laser system and the Army's DE-SHORAD program are years from full operational capability. Until affordable kill mechanisms are fielded at scale, the U.S. military is trapped in a lose-lose dynamic: expend expensive interceptors and go bankrupt, or withhold them and accept hits. Adversaries understand this arithmetic perfectly and are building their strategies around it.

Since November 2023, Yemen's Houthi forces have employed UAVs, cruise missiles, and ballistic missiles in hundreds of attacks on maritime shipping and land-based targets in the Red Sea and beyond. What makes this a structural shift rather than a temporary crisis is that the Houthis have transitioned from importing complete missile systems to manufacturing them domestically. In 2024 and 2025, more than 80% of items seized en route to the Houthis were raw materials and manufacturing components rather than finished weapons, indicating that the group has developed indigenous production capabilities for ballistic missiles and drones with Iranian technical assistance. This matters because it represents a fundamental change in missile proliferation dynamics. Traditional nonproliferation regimes like the Missile Technology Control Regime (MTCR) focus on preventing the transfer of complete missile systems and major subsystems. But when a non-state actor can manufacture missiles from commercially available materials and components, export controls become far less effective. The Houthis possess Scud-B and Scud-C variants, North Korean Hwasong derivatives, Tochka missiles, and Iranian-designed systems -- a diverse arsenal that a non-state group was never supposed to be able to field. The economic impact has been enormous. Houthi attacks on Red Sea shipping disrupted 12-15% of global trade that transits the Suez Canal, forcing vessels on costly detours around the Cape of Good Hope, driving up marine insurance premiums by 10-20x in the region, and adding weeks of transit time to global supply chains. The multinational naval coalition assembled to respond cost billions of dollars and consumed hundreds of interceptors that cost orders of magnitude more than the drones and missiles they were shooting down. The structural reason this proliferation persists is that Iran gains strategic leverage from transferring missile technology to proxy groups at minimal cost. For Iran, enabling Houthi missile capability creates a threat to Saudi Arabia, Israel, and global shipping that forces the U.S. and allies to expend enormous resources on defense -- a classic asymmetric strategy. The international community has no enforcement mechanism to stop this transfer: UN arms embargoes on the Houthis are routinely violated, the maritime interdiction campaign captures only a fraction of smuggled materials, and diplomatic pressure on Iran has failed to halt the flow. The precedent is alarming. If a non-state actor in Yemen can field an arsenal of ballistic missiles, cruise missiles, and long-range drones capable of threatening international shipping and striking targets 2,000 km away, other non-state actors with state sponsors will follow the same playbook. The era when ballistic missiles were exclusively the province of nation-states is over.

The New START treaty expired on February 5, 2026, without a replacement or extension, marking the first time since 1972 that no legally binding agreement limits the nuclear arsenals of the United States and Russia. Russia had offered an informal one-year extension of force limits in late 2025, but the Trump administration -- seeking a broader deal that would include China -- never formally responded. On February 4, 2026, Russia's Ministry of Foreign Affairs declared the treaty's obligations no longer binding. The collapse of New START is the final stage in the systematic dismantling of the arms control architecture: the ABM Treaty ended in 2002, the INF Treaty collapsed in 2019, and now the last remaining bilateral nuclear constraint is gone. This matters because arms control treaties are not just about limiting warhead counts -- they provide transparency, predictability, and verification mechanisms that reduce the risk of miscalculation. Under New START, both sides conducted up to 18 on-site inspections per year of each other's nuclear facilities. Those inspections provided direct intelligence about the other side's nuclear posture that no satellite or signals intelligence can replicate. Without them, both nations must rely on national technical means that are inherently less precise, increasing the risk that ambiguous intelligence is misinterpreted as a threat. The immediate practical consequence is that both the U.S. and Russia are now free to deploy more strategic nuclear weapons without any legal constraint. Russia has already indicated it will determine its force structure based on 'the evolving strategic environment,' a diplomatic signal that uploading additional warheads onto existing delivery systems is under consideration. The U.S. is simultaneously modernizing all three legs of its nuclear triad -- new ICBMs, new submarines, new bombers, new warheads -- at a projected cost of $1.5-2 trillion over 30 years. The structural reason arms control collapsed is the emergence of a tripolar nuclear dynamic that bilateral treaties cannot address. China is rapidly expanding its nuclear arsenal from approximately 500 warheads toward an estimated 1,000-1,500 by 2035, and has refused to participate in arms control negotiations, arguing its arsenal is too small to warrant limits. Russia will not accept constraints that do not account for Chinese capabilities on its border. The U.S. wants a trilateral framework but has no mechanism to compel Chinese participation. This three-body problem in nuclear diplomacy has no precedent and no obvious solution. The risk is a return to an open-ended nuclear arms competition among three major powers, with no verification, no transparency, and no agreed rules -- the most dangerous nuclear environment since the early 1960s before the first arms control agreements were reached.

The National Nuclear Security Administration (NNSA) is required to produce 80 plutonium pits per year by 2030 to support the W87-1 warhead for the Sentinel ICBM and other modernization programs. As of 2025, Los Alamos National Laboratory has produced its first 'diamond-stamped' pit but is nowhere near the 30-per-year interim target, let alone the 80-per-year goal. The second production site at the Savannah River Site in South Carolina will not have its main plutonium processing facility ready until 2035 at the earliest, with costs potentially exceeding $22 billion. The Government Accountability Office has reported that NNSA has no credible integrated master schedule or comprehensive cost estimate linking the two production sites. This matters because plutonium pits are the fissile cores of nuclear warheads -- the component that actually produces the nuclear yield. Without new pits, the United States cannot produce new warheads, and existing pits in the current stockpile are aging. The W87-1 warhead program, estimated at $15.9 billion, and the W80-4 warhead for the Long-Range Standoff Weapon, at $13 billion, both depend on pit production that does not yet exist at the required scale. Multi-year delays have already forced NNSA to resort to using recycled pits from dismantled warheads for some new weapons -- a stopgap that cannot sustain the full modernization program. The deeper problem is that the United States effectively abandoned large-scale pit production after the Rocky Flats Plant near Denver was shut down in 1989 due to environmental contamination. For over 35 years, the nation has relied on a small number of pits produced at Los Alamos in a facility not designed for mass production. Rebuilding this capability means constructing new facilities, training a specialized workforce in plutonium metallurgy, and navigating environmental regulations and community opposition -- all of which take decades, not years. The structural persistence of this problem reflects a fundamental tension in nuclear weapons policy: political leaders want a modernized deterrent but are unwilling to absorb the decades-long lead times and enormous costs required to rebuild production infrastructure that was dismantled a generation ago. Every administration since the 1990s has acknowledged the pit production gap; none has delivered a solution on schedule or on budget. The result is that the credibility of the entire nuclear modernization program -- estimated at $1.5-2 trillion over 30 years -- rests on an industrial capability that does not yet exist.

In 2025, officials from Nammo -- a major NATO solid rocket motor manufacturer -- discovered that a chemical company producing an essential propellant ingredient for one of its rocket motors was going out of business, with no alternative supplier anywhere in the world. This is not an isolated incident; it exemplifies a systemic fragility in the defense supply chain where specialty chemical producers, nozzle manufacturers, and propellant ingredient suppliers operate as irreplaceable single points of failure. Certain rocket motor nozzles require 7-10 months of lead time to source, and the loss of any single vendor can halt production of entire missile systems. This matters because solid rocket motors are the propulsion system for virtually every U.S. tactical and strategic missile: Patriot PAC-3, THAAD, SM-3, Stinger, Javelin, GMLRS, ATACMS, and the Minuteman III ICBM itself. If a sole-source supplier for a critical chemical or component goes bankrupt or ceases production, there is no quick substitution. Requalifying a new vendor for a propellant ingredient requires extensive testing, often taking 2-5 years and costing tens of millions of dollars, because any change to propellant chemistry affects burn rate, thrust profile, and warhead safety. During that requalification period, missile production stops. The demand surge has made the problem acute. Global conflicts in Ukraine, the Middle East, and potential Pacific contingencies have driven missile expenditure far beyond Cold War planning assumptions. The U.S. and allies are trying to simultaneously replenish stocks depleted by Ukraine aid, backfill interceptors used in Middle East operations, and build inventories for a potential Indo-Pacific conflict -- all while the industrial base was sized for peacetime procurement. The structural cause is decades of defense industry consolidation that eliminated redundancy. During the post-Cold War drawdown, the number of solid rocket motor producers collapsed. Small specialty chemical companies that serve defense as a side business have no incentive to maintain production lines for low-volume military orders when commercial markets offer better margins. The defense sector represents a tiny fraction of the overall chemicals market, giving it no leverage to prevent supplier exits. Congress has begun responding: the 2025 reconciliation bill included $200 million for the solid rocket motor industrial base, $400 million for emerging SRM makers, $42 million for second sources of large-diameter motors for hypersonic missiles, and $100 million for a second source of Navy air defense motors. The Department of Defense also approved $32.7 million in targeted investments in September 2025. But rebuilding an industrial base that took 30 years to hollow out will not happen in a single budget cycle.

Hypersonic weapons -- missiles traveling above Mach 5 that can maneuver throughout their flight -- present thermal signatures 10 to 20 times fainter than traditional ballistic missiles, rendering current U.S. geostationary satellite-based detection systems largely ineffective. Unlike ballistic missiles that follow predictable parabolic trajectories after boost phase, hypersonic glide vehicles maintain powered, maneuverable flight at altitudes below the detection threshold of most terrestrial radars, leaving defenders with minutes or even seconds of warning rather than the 20-30 minutes provided by traditional ICBM detection. This matters because the entire U.S. missile defense architecture was built around the assumption that incoming threats follow ballistic trajectories. Ground-based radars, space-based infrared satellites, and command-and-control systems all rely on predicting where a missile will be based on its boost-phase trajectory. Hypersonic weapons break this assumption entirely. They compress decision timelines from hours to minutes, meaning that by the time a hypersonic threat is detected, there may be insufficient time to observe, orient, decide, and act -- the fundamental OODA loop that military commanders depend on. A successful hypersonic strike against a carrier group, forward base, or command node could be accomplished before any defensive response is possible. China has deployed the DF-ZF hypersonic glide vehicle and tested the DF-17 medium-range system. Russia has fielded the Avangard hypersonic glide vehicle on its ICBMs and deployed the Kinzhal air-launched hypersonic missile in combat in Ukraine. North Korea has tested its own hypersonic prototypes. The proliferation is accelerating while the defensive gap remains open. The structural reason this problem persists is that detecting and tracking hypersonic weapons requires a fundamentally different sensor architecture -- a proliferated constellation of satellites in low Earth orbit rather than a small number of exquisite satellites in geostationary orbit. The Hypersonic and Ballistic Tracking Space Sensor (HBTSS) program and the Space Development Agency's Tracking Layer are being developed to fill this gap, but fielding a full operational constellation requires launching hundreds of satellites, integrating them with existing command infrastructure, and developing fire-control-quality tracking data -- a process that will take years and billions of dollars. Until this new sensing architecture is operational, the U.S. faces a window of vulnerability where adversary hypersonic weapons can potentially penetrate existing defenses. A March 2025 MDA and Navy test demonstrated that HBTSS data could detect and track a maneuvering hypersonic target in a simulated engagement, but moving from a successful test to a deployed, reliable, global tracking capability is a multi-year journey that adversaries are racing to exploit.

During the June 2025 conflict involving Iran, the United States expended over 150 THAAD interceptors and approximately 80 SM-3 interceptors in a single defensive operation -- depleting roughly 25% of the total U.S. THAAD stockpile in days. As of December 2025, the Missile Defense Agency's inventory stood at 534 THAAD interceptors and 414 SM-3s, with a backlog of 100 THAAD missiles procured between fiscal years 2021 and 2024 but not yet delivered. No new THAAD interceptors are expected until April 2027. This matters because missile defense interceptors are the finite resource that stands between adversary missile strikes and defended populations and military assets. When stockpiles drop below a critical threshold, the U.S. loses the ability to simultaneously defend allies in multiple theaters. A THAAD battery deployed to the Middle East means one fewer battery available for the Korean Peninsula or Guam. The June 2025 expenditure demonstrated that a single regional conflict can consume a quarter of the nation's highest-tier missile defense inventory in under a week, a burn rate that current production cannot sustain. The asymmetric cost problem compounds this. Each THAAD interceptor costs approximately $12-15 million. Each PAC-3 MSE interceptor costs $3-5 million. Adversaries can launch ballistic missiles and drones costing a fraction of these amounts, creating a ratio where defenders spend $15-35 for every $1 the attacker spends. Navy officials described this reliance on top-tier interceptors against low-cost threats as 'unsustainable' by mid-2024, yet no affordable alternative is fielded at scale. The structural cause is a production base designed for peacetime procurement rates, not wartime consumption. THAAD interceptor production was historically capped at 96 per year. Even after the January 2026 agreement with Lockheed Martin to quadruple production to 400 per year, it will take years to ramp up and replenish depleted stocks. The industrial base for solid rocket motors, guidance seekers, and kill vehicles involves specialized facilities with multi-year lead times for expansion. The strategic consequence is that adversaries like Iran, North Korea, and China can observe U.S. interceptor expenditure in real time and calculate how many salvos it would take to exhaust American missile defense capacity. This creates a perverse incentive: the more the U.S. uses its defenses, the weaker it becomes, and adversaries know exactly how to exploit this arithmetic of attrition.

The U.S. Air Force's LGM-35A Sentinel program, intended to replace the 50-year-old Minuteman III ICBM, has seen its total acquisition cost balloon from $77.7 billion to $140.9 billion -- an 81% overrun triggered by the discovery that Cold War-era silos cannot support a new missile through the 2070s and that legacy communications cabling is too degraded to interface with modern systems. The Air Force now acknowledges that entirely new silos and infrastructure must be built, and initial operational capability has slipped to the early 2030s. This matters because the Minuteman III was designed for a 10-year service life and has been in operation for over 50 years. Every year the replacement slips, the Air Force must sustain a weapon system whose ground electrical subsystems -- diodes, resistors, capacitors -- may degrade to unacceptable levels at any time. Flight-test components are in limited supply, meaning the service cannot indefinitely verify the missile actually works. A September 2025 GAO report found that extending Minuteman III to 2050 is technically feasible but carries significant risk of undetected degradation in components that have no modern equivalent. The structural reason this problem persists is the defense acquisition system's inability to accurately scope infrastructure costs before committing to a program. The original Sentinel cost estimate assumed Cold War silos could be reused with minor upgrades -- a fundamentally flawed assumption that was not challenged until billions had already been obligated. The program is now in a Nunn-McCurdy critical cost breach, the most severe category of overrun in defense procurement, yet cancellation would leave the U.S. with no path to replacing an aging nuclear deterrent. The result is a procurement trap: too expensive to continue as planned, too critical to cancel. Meanwhile, the workforce that maintains Minuteman III faces its own crisis. Maintainers and missile-field security personnel are in short supply, and the extended timeline means the Air Force must retain specialized skills for a system that was supposed to be retired years ago. Recruitment into nuclear missile maintenance is notoriously difficult given the remote basing locations across Montana, Wyoming, North Dakota, Colorado, and Nebraska. The net effect is that the land-based leg of the U.S. nuclear triad -- 400 deployed ICBMs across roughly 45,000 square miles -- is caught between a decaying present system and a future system that keeps getting more expensive and further away. Every dollar of Sentinel overrun is a dollar not available for other modernization priorities, and every year of delay is a year of compounding risk to the existing deterrent.

The Planning, Programming, Budgeting, and Execution (PPBE) system requires the DoD to plan its budget 2-3 years before funds are available and obligate most funds within fixed fiscal year windows. A technology opportunity that emerges in March 2025 cannot receive funding until the FY2028 budget at the earliest — and that is only if a service prioritizes it in its Program Objective Memorandum (POM) build, which happens once per year. By the time money is available, the technology window may have closed, a competitor may have fielded it first, or the startup that developed it may have gone bankrupt. This budgeting rigidity is the single largest structural barrier to DoD technology adoption. Every other reform — OTA contracts, DIU prototyping, SBIR transitions — eventually hits the PPBE wall. A program manager who wants to buy a proven commercial AI system today cannot do so because their budget line funds a different capability and reprogramming requires congressional notification for any amount over $20 million. The reprogramming process itself takes 6-12 months and is limited to a small percentage of each appropriation. The net effect is that the DoD's budget is a snapshot of priorities from 3 years ago, not a response to current threats. PPBE persists because it serves Congress's constitutional power of the purse. Congressional appropriators want to control how every dollar is spent, and PPBE gives them that control through detailed budget line items and strict rules about moving money between accounts. The defense committees' staffs — who are the real power behind military budgeting — have built their expertise around PPBE and resist changes that would reduce their oversight granularity. The system also benefits the services' programming shops (N81, A8, AF/A8), which derive institutional power from controlling the POM process. The Section 809 Panel, the PPBE Reform Commission (established by FY2022 NDAA), and numerous defense analysts have recommended fundamental PPBE reform. The PPBE Commission's 2024 report proposed shifting to portfolio-based budgeting with more reprogramming flexibility, but implementation requires Congress to voluntarily cede appropriation-level control — something appropriators have historically refused to do. The few workarounds that exist (rapid acquisition funds, Commander's Emergency Response Program-style authorities) are small, temporary, and constantly at risk of being clawed back by appropriators who view them as end-runs around congressional oversight.

The DoD procures millions of electronic components annually for weapons systems, many of which pass through opaque supply chains with components manufactured in China, assembled in Southeast Asia, and sold through multiple distributors before reaching a defense contractor. Verifying that a microchip or circuit board is authentic and free of hardware trojans or backdoors is technically difficult and rarely done comprehensively. The Government Accountability Office has repeatedly found counterfeit electronic parts in defense systems, including in missile defense radars and aircraft avionics. Counterfeit or compromised hardware in weapons systems is not a theoretical risk — it is a demonstrated one. A counterfeit transistor that fails under stress could cause a radar to go blind at the worst possible moment. A hardware trojan implanted in a processor could exfiltrate classified data or disable a system on command. The entire premise of deterrence depends on adversaries believing that American weapons systems will work as designed. Supply chain compromise undermines that certainty in ways that may not be discovered until combat. The problem persists because the global semiconductor supply chain was optimized for cost, not security. American defense contractors source from the same global supply chains as commercial manufacturers because domestic production of many component types does not exist. The CHIPS Act is investing $52 billion in domestic semiconductor manufacturing, but this addresses cutting-edge logic chips (14nm and below) — the vast majority of defense-critical components are mature-node chips (28nm and above) that are not covered. Even with domestic fabrication, assembly, testing, and packaging (ATP) overwhelmingly occurs in Asia. The DoD's current approach relies on the Trusted Foundry Program and DMEA (Defense Microelectronics Activity), but these cover only a tiny fraction of the components the DoD needs. The Trusted Foundry Program has been criticized as expensive, slow, and technologically lagging. More fundamentally, no testing methodology can guarantee the absence of a hardware trojan in a complex integrated circuit — the problem is computationally intractable for modern chip designs with billions of transistors. The DoD is left relying on supply chain risk management processes that reduce but cannot eliminate the risk.

Before any software system can be deployed to warfighters, it must pass through operational test and evaluation (OT&E) conducted by the Director of Operational Test and Evaluation (DOT&E), an independent office that reports directly to Congress. OT&E was designed for hardware systems — testing whether a missile hits its target or an aircraft meets performance specifications. Applied to software, OT&E imposes a testing paradigm that assumes software is a finished product to be validated, rather than a continuously evolving system. A single OT&E cycle takes 12-24 months and must be repeated for each major software release. The consequence is that warfighters receive software updates years after they are developed. A bug fix written in January 2024 might not reach the fleet until 2026 because it must be bundled into a major release, scheduled for OT&E, tested over months, findings adjudicated, and then approved for deployment. In the commercial world, software companies deploy multiple times per day. In the DoD, deployments happen once or twice per year. The gap between commercial and military software delivery speed is not narrowing — it is widening, because commercial practices are accelerating while DoD processes remain static. OT&E persists in its current form because DOT&E has strong institutional incentives to maintain its role. It is a congressionally mandated office that justifies its existence by finding problems in weapons systems. If DOT&E endorsed continuous delivery with automated testing, it would be endorsing a model that reduces the need for its own independent evaluation. Congressional defense committees value DOT&E as an independent check on service acquisition programs and are reluctant to reduce its authority. The Goldwater-Nichols Act of 1986, which established DOT&E's independence, would need to be amended — a heavy legislative lift. The DoD's own DevSecOps initiatives (Platform One, Black Pearl, Party Bus) have demonstrated that continuous integration and automated testing can deliver software rapidly while maintaining security. But these platforms operate in a gray zone — they have not fully replaced OT&E, and programs that use them still face pressure to undergo traditional testing for major milestones. The cultural shift from 'test at the end' to 'test continuously' requires not just new tools but a fundamental change in how Congress oversees software acquisition.

As of 2024, the Defense Counterintelligence and Security Agency (DCSA) had a backlog of over 300,000 pending background investigations. A Top Secret clearance takes a median of 8-12 months to adjudicate, and Special Access Program (SAP) access adds another 3-6 months on top. During this waiting period, the applicant cannot work on classified programs — they either sit idle doing unclassified busywork at their employer's expense or take a different job entirely and never return to defense work. The clearance backlog is a hidden tax on the entire defense industrial base. Companies must carry the cost of cleared personnel who cannot bill to contracts while waiting for reciprocity or upgrade processing. Small companies and startups are disproportionately affected because they cannot absorb this cost. A defense startup that hires an engineer in January and cannot put them on a classified contract until November may burn through $150,000 in salary before generating any revenue from that employee. This is a direct barrier to entry that protects incumbent contractors and stifles competition. The problem exists because the clearance process was designed in the 1950s for a workforce that changed jobs infrequently. The investigation model — sending federal agents to interview neighbors, former employers, and references in person — does not scale to a mobile, digital workforce. DCSA's Trusted Workforce 2.0 initiative promises continuous vetting using automated data checks, but implementation has been repeatedly delayed. The transition from OPM to DCSA itself caused years of disruption after the 2015 OPM data breach that compromised 21.5 million clearance records. Reciprocity failures between agencies compound the backlog. A contractor with a DoD Top Secret clearance who moves to an Intelligence Community program must undergo a substantially new investigation, even though the same information is being verified. Despite executive orders mandating reciprocity, each agency maintains its own adjudication standards and systems that do not interoperate. The bureaucratic incentive is clear: each agency's security office justifies its existence by maintaining independent control over who it trusts.

The five major defense primes (Lockheed Martin, RTX, Northrop Grumman, Boeing, and General Dynamics) routinely deliver weapons systems with proprietary data formats, closed interfaces, and contractual restrictions that prevent the government from modifying, maintaining, or integrating the systems without paying the original contractor. When the DoD wants to add a new sensor to an aircraft or update a ship's combat system, it must go back to the prime — often sole-source — at whatever price the prime sets. Technical data rights that should belong to the government are routinely negotiated away during contracting. This vendor lock-in costs the DoD billions annually in sustainment costs and, more importantly, prevents rapid capability integration. When a new threat emerges, the warfighter cannot quickly integrate a new electronic warfare pod or software update because the platform's interfaces are proprietary. The speed of capability delivery — the single most important factor in modern warfare — is held hostage by a contractor's business model. In Ukraine, forces modify commercial drones in days. The US military takes months to modify its own aircraft because it does not control the interfaces. The problem persists because of a fundamental misalignment of incentives in the defense contracting model. Primes bid low to win development contracts knowing they will make their profits in decades of sustainment. Proprietary lock-in is the mechanism that guarantees those sustainment profits. The DoD's contracting workforce is trained to evaluate lowest price technically acceptable (LPTA) bids, not total lifecycle cost including switching costs. Even when contracting officers understand the lock-in risk, they face pressure to award contracts quickly and lack the technical expertise to specify open interface requirements precisely enough to be enforceable. Congress has mandated modular open systems approaches (MOSA) since the FY2017 NDAA, and the DoD has published open architecture standards. But compliance is largely performative — primes deliver systems that technically meet open standards requirements while ensuring that the truly valuable integration logic remains proprietary. The government lacks the organic engineering talent to audit whether a delivered system is genuinely open or merely superficially compliant.

A GS-13 software engineer at the Department of Defense earns $90,000-$120,000 in total compensation. The same engineer at Google, Meta, or a well-funded defense startup earns $300,000-$500,000. The DoD cannot offer stock options, signing bonuses above $25,000 (requiring senior approval), or competitive remote work policies. The result is that the DoD's organic software workforce is aging, shrinking, and increasingly unable to evaluate, manage, or build the software systems that determine military superiority. The talent gap does not just mean fewer government developers — it means the government cannot be an intelligent buyer of software from contractors. When the program office lacks engineers who understand modern software architecture, they cannot write good requirements, evaluate contractor proposals, or identify when a contractor is delivering poor quality work. This is why the DoD pays $200-400 per hour for contractor software engineers doing work that costs $100 per hour in the commercial sector. The government's inability to hire technical talent makes it a worse customer, which increases costs, which consumes budget that could fund more capability. The structural cause is the General Schedule (GS) pay system, established by the Classification Act of 1949. GS pay bands are based on the premise that government work is comparable to private sector work at equivalent education and experience levels — a premise that was approximately true in 1949 but is wildly false for software engineers in 2025. Congress has authorized special pay authorities (like cyber excepted service), but each authority is narrowly scoped, requires extensive justification, and has been used sparingly because HR offices lack the technical knowledge to classify software roles appropriately. The deeper issue is cultural. The DoD's human capital system values credentials (degrees, clearances, years of service) over demonstrated capability. A 22-year-old who can architect distributed systems at scale will be hired as a GS-7 because they lack a master's degree and have only two years of experience. Meanwhile, a 45-year-old with a master's and 20 years of irrelevant experience will be hired as a GS-14. The system selects for tenure over talent, and no amount of special pay authorities fixes a classification system that cannot recognize what good software engineering looks like.

A defense technology startup that wins a Small Business Innovation Research (SBIR) Phase I contract receives $50,000-$250,000 and a proof of concept. Phase II provides $500,000-$1.5 million for a prototype. But between Phase II completion and an actual production contract (Phase III), there is a 2-3 year gap with zero government funding. The startup must survive on venture capital while navigating the byzantine process of getting a program of record to adopt their technology. Most do not survive. This valley of death means the DoD systematically funds innovation and then watches it die before fielding it. Taxpayers pay for R&D that never reaches warfighters. The startups that do survive are those with enough VC funding to burn cash for years — selecting not for the best technology but for the best-funded. Anduril survived the valley of death because Palmer Luckey invested $200 million of personal wealth. Most founders cannot do that. The result is a defense industrial base that remains dominated by the same five primes (Lockheed, Raytheon, Northrop, Boeing, General Dynamics) who know how to navigate procurement but have weak incentives to innovate. The problem persists because no single organization in the DoD owns the transition from prototype to production. SBIR program managers fund R&D. Acquisition program managers buy production systems. These are different people in different organizations with different budgets and different incentive structures. The acquisition PM has zero incentive to take a risk on a startup's unproven technology when they can buy a proven (if outdated) system from a prime and not get blamed if it works poorly. Innovation adoption is a career risk with no career reward. Congress has attempted fixes — the SBIR Phase III mandate, the Rapid Innovation Fund, and various OTA (Other Transaction Authority) pathways — but they treat symptoms rather than the structural cause. The fundamental issue is that DoD budgets are allocated to programs of record years in advance, and inserting a new technology into an existing program requires re-baselining the program, which triggers oversight reviews, which no PM wants to initiate.

Software developers working on classified defense programs must use air-gapped networks (SIPRNet, JWICS, or Special Access Program facilities) that lack access to the modern internet, open-source libraries, Stack Overflow, GitHub, package managers, and AI coding assistants. A developer who writes 50 features per sprint in a commercial environment produces roughly 10 in a classified one. The machines are often outdated (approved hardware lags commercial availability by 3-5 years), and getting new tools approved through the Authority to Operate (ATO) process takes 12-18 months. This productivity collapse has a devastating cascading effect. Defense software projects already face talent shortages — when each developer is 80% less productive, the effective workforce shrinks by another 5x. Programs fall behind schedule, which triggers cost overruns, which triggers congressional scrutiny, which triggers more oversight and documentation requirements, which further reduces developer productivity. It is a death spiral. The warfighter receives inferior software years late, while adversaries with no such constraints iterate rapidly. The reason classified computing environments remain so hostile to developers is the accreditation model. The Risk Management Framework (RMF) process, managed by DISA and enforced by each organization's authorizing official, requires every piece of software — down to individual libraries and updates — to be documented, scanned, and approved before installation. The authorizing official faces severe personal consequences for a security breach but zero consequences for slow approvals. The rational individual-level decision is to approve nothing unless absolutely forced to, creating a bureaucratic immune system that attacks productivity. Structurally, the classification system itself is the problem. Programs classify entire codebases at the highest level of any component, even when 90% of the code is unclassified algorithms, UI logic, and infrastructure. There is no practical mechanism for developers to work on unclassified portions in a modern environment and then integrate with classified components in a secure facility. Cross-domain solutions exist but are expensive, rare, and themselves subject to multi-year accreditation timelines.

The International Traffic in Arms Regulations (ITAR) treat defense technology exports to the UK, Australia, and other Five Eyes allies with nearly the same bureaucratic burden as exports to non-allied nations. A State Department export license for a technical data package can take 4-16 months to process. Even a single ITAR-controlled component in a multinational system forces the entire program into the ITAR regime, causing allied partners to actively design around American technology to avoid contaminating their supply chains. The real-world consequence is that America's closest allies are choosing non-American defense systems specifically to avoid ITAR. Australia chose the French Naval Group (later switched to UK design) for its submarine program partly due to ITAR frustrations. The UK's Tempest sixth-generation fighter program explicitly sought to minimize US components. When allies avoid American technology, interoperability suffers — the core advantage of the Western alliance in combat is the ability to share data, communicate, and fight as an integrated force. ITAR is actively undermining this. The structural reason ITAR persists in this form is that the State Department's Directorate of Defense Trade Controls (DDTC) is chronically understaffed and has zero incentive to approve licenses faster. Each approval carries career risk for the reviewing officer (a leak becomes a scandal), while delays carry no personal consequence. Congress passed the ITAR exemption for Australia and the UK under AUKUS in the FY2024 NDAA, but implementation has been glacially slow because DDTC must rewrite decades of regulatory guidance. Defense contractors also quietly benefit from ITAR because it creates switching costs — once an ally is locked into an American platform, they cannot easily share technical data with third-party maintainers. The deeper issue is that ITAR was written during the Cold War when the primary concern was preventing Soviet acquisition of American technology. The threat model has changed — China steals technology primarily through cyber espionage and academic collaboration, not through allied defense cooperation — but the regulatory framework has not adapted. The 2020 ITAR reform moved some items to the Commerce Control List, but the vast majority of defense-relevant technology remains under State Department control with its slower, more risk-averse process.

The Department of Defense's software acquisition process averages 5-7 years from requirements definition to initial operational capability. During this time, the threat landscape shifts dramatically — adversaries deploy new electronic warfare systems, cyber capabilities, and autonomous weapons on timelines measured in months, not years. By the time a software system is fielded, the requirements it was built against are obsolete. This matters because software is now the decisive factor in modern warfare. The F-35's 8 million lines of code determine its combat effectiveness more than its airframe. When software delivery is slow, warfighters go into combat with outdated capabilities. In Ukraine, both sides iterate on drone software weekly — the side that updates faster survives. A 5-year software cycle in that context is not just inefficient, it is lethal. The problem persists because DoD procurement was designed for hardware: tanks, ships, and aircraft that take decades to build and remain relevant for decades. The entire acquisition workforce — contracting officers, program managers, test organizations — is trained in waterfall hardware processes. The Federal Acquisition Regulation (FAR) and Defense Federal Acquisition Regulation Supplement (DFARS) encode these assumptions into law. Even when programs try agile approaches, they must still navigate Milestone Decision Authority reviews, operational test and evaluation, and authority-to-operate processes that assume software is a static deliverable rather than a continuously evolving product. Congress compounds the problem through the Planning, Programming, Budgeting, and Execution (PPBE) system, which requires funding to be planned 2-3 years in advance and obligated within fixed fiscal year windows. Software teams cannot respond to emerging needs because their budgets were locked years ago for different requirements. The Section 809 Panel identified over 1,300 regulatory barriers to faster acquisition, yet most remain in place because changing them requires coordinated action across Congress, OSD, and the services.

The systems that control the United States' nuclear arsenal — including the command, control, and communications (NC3) infrastructure for detecting attacks, conveying presidential orders, and executing launch sequences — rely on computing hardware and software that in some cases dates to the 1970s. A 2016 GAO report found that the Department of Defense's Strategic Automated Command and Control System (SACCS) still used 8-inch floppy disks and ran on an IBM Series/1 computer from the 1970s. While upgrades have been initiated, the modernization program is decades behind schedule and over budget. This matters because these are the systems that prevent accidental nuclear war and ensure that if deterrence fails, the response is authorized, proportional, and directed at the right targets. When the hardware that carries presidential nuclear orders uses technology older than the internet, every component is a potential single point of failure. In 2010, a hardware failure at F.E. Warren Air Force Base caused 50 Minuteman III ICBMs to go offline simultaneously for 45 minutes, during which the missiles could not be monitored or launched. The Air Force initially could not determine whether the outage was a cyberattack, a hardware failure, or an adversary action. The cascading risk is that an NC3 failure during a crisis could be misinterpreted by either side. If the US loses communication with a missile wing during a period of heightened tension, decision-makers must determine whether the outage is technical or the result of an adversary's first strike — with minutes to decide. Conversely, if an adversary detects the US scrambling to restore nuclear communications, they might interpret the activity as preparation for a strike. The aged infrastructure creates ambiguity at precisely the moments when clarity is most critical to preventing nuclear use. This problem persists because nuclear modernization competes for funding with conventional military programs that have more visible constituencies and more immediate operational needs. The NC3 modernization program is estimated to cost $100 billion over 30 years, but the systems are classified, the workforce is small and specialized, and there is no commercial equivalent to leverage. The engineers who built the original systems have retired or died, and documentation is incomplete. Each component upgrade risks introducing new failure modes into a system where failure could mean civilization-ending consequences, creating a bureaucratic paralysis where the risk of changing the system is perceived as comparable to the risk of keeping it.

Autonomous drone swarms — coordinated groups of dozens to hundreds of small unmanned aerial systems operating on shared algorithms without individual human control — have outpaced every existing defensive countermeasure in both testing and combat. During the 2019 attack on Saudi Aramco's Abqaiq facility, roughly 18 drones and 7 cruise missiles penetrated one of the most heavily defended airspaces in the Middle East, temporarily knocking out 5% of global oil supply. By 2024, Ukraine and Russia routinely launched drone attacks involving 50-100+ drones per wave, with defenders unable to intercept more than a fraction. This matters because the economics are catastrophically asymmetric. A commercial-grade FPV drone costs $500-2,000. A single Patriot interceptor missile costs $4 million. Even a MANPADS like Stinger costs $120,000. Defending against a swarm of 100 drones with kinetic interceptors costs orders of magnitude more than the attack itself, and the defender runs out of ammunition before the attacker runs out of drones. This inverts the traditional advantage that defense held over offense in conventional warfare. The operational consequence is that any high-value fixed target — airbases, command centers, logistics depots, ships in port, critical infrastructure — is now vulnerable to destruction by an adversary with access to commercial drone technology and basic autonomous coordination algorithms. The barrier to building a militarily significant drone swarm capability has fallen to the level of a well-funded non-state actor. Houthi forces in Yemen have demonstrated this repeatedly against Saudi, UAE, and commercial shipping targets using modified commercial components. This problem persists because counter-swarm technology development is fragmented across dozens of programs with no unified architecture. Electronic warfare (jamming) works against remotely piloted drones but not against fully autonomous swarms navigating by onboard vision. Directed energy works against individual drones but cannot slew fast enough for simultaneous multi-axis threats. Kinetic systems work but at unaffordable cost ratios. The defense acquisition timeline for fielding integrated counter-swarm systems is 5-10 years; the timeline for an adversary to scale drone production is months. The offense-defense balance has fundamentally shifted, and procurement bureaucracies have not adapted.

Despite decades of investment exceeding $40 billion since the 1980s Strategic Defense Initiative, directed energy weapons (DEWs) — including high-energy lasers and high-powered microwaves — remain unable to operate effectively beyond short-range point defense roles due to fundamental physics constraints that no amount of engineering has overcome. Atmospheric absorption, scattering, thermal blooming (where the laser heats the air and defocuses itself), and the need for sustained beam dwell time on target all limit effective engagement ranges to a few kilometers in realistic battlefield conditions with dust, humidity, and smoke. This matters because military planners and defense contractors have repeatedly promised that DEWs would revolutionize warfare by providing unlimited magazines at near-zero marginal cost per shot. The US Navy's deployment of the HELIOS laser on USS Preble and the Army's DE-SHORAD program are real capabilities, but they are effective only against small drones and rockets at short range in favorable atmospheric conditions. They cannot engage aircraft, cruise missiles, or ballistic threats at the ranges needed to protect against peer adversary weapons. The mismatch between promise and performance has diverted billions from proven kinetic interceptor programs that work today. The operational consequence is that forces planning to rely on DEWs for defense against drone swarms and missile salvos may find themselves critically underprotected. A 300-kilowatt laser that works perfectly against a single drone at 2 km in clear desert air may be overwhelmed by a swarm of 50 drones or rendered ineffective by battlefield smoke, rain, or fog. The power generation requirements — a 300 kW laser requires roughly 1 MW of electrical power including cooling — limit DEWs to large ships and fixed installations, excluding the mobile ground forces that most need counter-drone protection. This problem persists because the physics constraints are fundamental, not engineering limitations waiting to be solved. You cannot engineer around atmospheric absorption at laser wavelengths. The defense industrial base has financial incentives to sustain DEW programs regardless of operational limitations, and the narrative of a "game-changing" technology is politically attractive to both military leaders seeking modernization budgets and politicians seeking innovation stories. The result is a recurring cycle: overpromise, underdeliver, rebrand, and request more funding.

Military precision strike doctrine relies on Collateral Damage Estimation (CDE) methodologies that systematically undercount civilian casualties by using outdated population density data, failing to account for sheltering populations, and applying blast models calibrated to open terrain rather than urban environments. A 2022 RAND study commissioned by the Pentagon found that actual civilian casualties from US airstrikes in Iraq and Syria were 5-10 times higher than official CDE predictions. The Airwars monitoring group documented over 13,000 civilian deaths from Coalition strikes in the campaign against ISIS, compared to the Coalition's acknowledged total of roughly 1,400. This matters because the entire legal and ethical framework for modern air warfare rests on the proportionality principle: a strike is lawful only if the expected military advantage outweighs the expected civilian harm. When the CDE methodology systematically undercounts expected civilian harm by a factor of 5-10x, commanders are approving strikes based on a proportionality analysis that is fundamentally wrong. Strikes that would be rejected as disproportionate under accurate casualty estimates are approved because the model says fewer civilians will die. The strategic consequence extends beyond individual strikes. Systematic undercounting allows political leaders and military institutions to maintain the narrative that precision warfare is clean and surgical, which lowers the political threshold for approving military operations. If the true civilian cost were acknowledged upfront, both domestic publics and international allies might withhold support. The gap between reported and actual casualties creates a credibility deficit that adversaries exploit for propaganda and that erodes trust in democratic institutions' claims about rules-based warfare. This persists because reforming CDE methodology would mean acknowledging that past strikes killed far more civilians than reported — creating legal liability, political embarrassment, and potential war crimes investigations. The institutional incentive is to maintain current models. Additionally, the classification of CDE inputs (population data, intelligence assessments, weapon effect models) prevents independent review. When the New York Times investigated Pentagon civilian casualty records in 2021, they found that internal assessments routinely dismissed credible reports of civilian harm without investigation.

The majority of Western precision-guided munitions — including JDAM kits, Excalibur artillery shells, and cruise missiles — depend on GPS signals for terminal guidance, and these signals are trivially jammed or spoofed by modern electronic warfare systems costing a fraction of the munitions they defeat. During the Ukraine war, Russia's GPS jamming capabilities rendered significant quantities of GPS-guided weapons ineffective: Ukrainian forces reported that JDAM-ER bombs provided by the US were missing targets by hundreds of meters due to Russian jamming, and the US had to rush development of upgraded anti-jam kits. This matters because the US and NATO have spent decades and hundreds of billions of dollars building precision strike doctrines around GPS-guided weapons. The assumption that these munitions will hit within meters of their aim point underpins war plans, collateral damage estimates, and rules of engagement. When GPS jamming degrades accuracy from 5 meters to 500 meters, a weapon designed for surgical strikes becomes an area weapon — the very thing precision guidance was supposed to eliminate. The operational consequence is that Western militaries face a scenario where their most numerous and affordable precision munitions become unreliable in exactly the environments where they are most needed — contested battlefields against near-peer adversaries with electronic warfare capabilities. This forces reliance on far more expensive alternatives like laser-guided munitions (which require clear weather and a designator aircraft in range) or inertial navigation (which drifts over distance), or accepting dramatically higher collateral damage. This problem persists because GPS dependency was baked into Western weapons procurement over three decades when the US enjoyed unchallenged electromagnetic spectrum dominance. The Pentagon's inventory contains hundreds of thousands of GPS-guided munitions that cannot be retroactively upgraded to resist jamming. Alternative navigation technologies — terrain contour matching, visual scene matching, quantum inertial navigation — exist in labs but are years from fielded production. Meanwhile, GPS jammers continue to get cheaper and more powerful: Russia deploys vehicle-mounted systems that can deny GPS across hundreds of square kilometers.

The international export control regime — centered on the Wassenaar Arrangement, the Missile Technology Control Regime (MTCR), and national systems like the US ITAR/EAR — consistently fails to prevent the transfer of critical offensive weapons technologies to adversarial states and non-state actors. A 2023 Royal United Services Institute (RUSI) study found that over 450 foreign-made components — including Western-manufactured microchips, sensors, and navigation modules subject to export controls — were recovered from Russian weapons systems used in Ukraine, including Kalibr cruise missiles and Orlan-10 drones. This matters because the entire theory of export controls is that denying adversaries access to key technologies will degrade their ability to build advanced weapons. When Russian cruise missiles striking Ukrainian cities contain US-made microprocessors that were supposedly controlled, the policy has failed at its most fundamental objective. The components found were not obscure dual-use items in gray areas — they included specific chips from Texas Instruments, Analog Devices, and Intel that are on sanctions lists. The downstream consequence is that export control failures enable adversaries to build precision-guided weapons at lower cost and higher volume than they could with purely indigenous components. Russia's ability to sustain cruise missile production throughout the Ukraine war, despite sanctions, demonstrates that procurement networks can route controlled components through intermediary countries — particularly in Central Asia, the UAE, and Turkey — faster than enforcement agencies can shut them down. This problem persists structurally because export controls were designed for a Cold War era with clear bloc boundaries and limited global trade networks. Today's supply chains involve dozens of intermediaries across multiple jurisdictions, and a $5 microchip can be rerouted through five countries before reaching its final destination. The US Bureau of Industry and Security (BIS) has fewer than 500 employees to monitor millions of export transactions annually. Wassenaar operates on voluntary compliance with no enforcement mechanism. The economic incentives for intermediary companies and countries to facilitate re-export far outweigh the penalties, which are rarely imposed.

Hypersonic weapons — missiles traveling at Mach 5 or faster that can maneuver unpredictably during flight — reduce the time available for human decision-makers to detect, assess, and respond to an incoming strike to as little as 6-15 minutes, and in some scenarios under 5 minutes for terminal-phase engagement decisions. Russia's Avangard hypersonic glide vehicle, China's DF-ZF, and the US's Long Range Hypersonic Weapon all travel at speeds that outpace existing early warning and decision-making architectures designed around the 25-30 minute flight time of traditional ICBMs. This compression of decision time matters because nuclear strategy has always relied on humans having enough time to verify warnings, consult with advisors, and make deliberate choices about whether and how to respond. The 1983 Soviet nuclear false alarm — where Lt. Col. Stanislav Petrov correctly identified a satellite warning as a malfunction rather than launching a retaliatory strike — required roughly 20 minutes of human judgment. With hypersonic weapons, that window shrinks to the point where automated systems may need to recommend or even execute responses before a human can meaningfully evaluate the situation. The cascading risk is that nations facing hypersonic threats will pre-delegate launch authority to lower-level commanders or integrate automated response systems, both of which dramatically increase the probability of accidental nuclear war. China has historically maintained a no-first-use policy partly because its leadership had confidence in its ability to absorb a first strike and retaliate. If hypersonic weapons threaten to destroy China's command-and-control infrastructure before leadership can authorize a response, China faces pressure to adopt launch-on-warning postures that make accidental escalation far more likely. This problem persists because hypersonic weapons provide genuine military advantages — they can penetrate existing missile defense systems — and no nation developing them is willing to negotiate limits when they believe the technology gives them strategic superiority. Arms control negotiations move at diplomatic speed while hypersonic programs advance at engineering speed. The absence of hypersonic weapons from any existing arms control framework (New START covers ICBMs but not conventional hypersonic systems) means there is no institutional mechanism even to discuss mutual restraint.

Nation-state cyberweapons developed for offensive military operations have repeatedly escaped their intended targets and caused billions of dollars in collateral damage to civilian infrastructure worldwide. The most notorious example is NotPetya, a cyberweapon attributed to Russian military intelligence (GRU) that was initially deployed against Ukrainian tax software in June 2017. The malware spread globally within hours, shutting down Maersk's shipping operations (costing $300 million), Merck's pharmaceutical manufacturing ($870 million), FedEx's TNT Express ($400 million), and countless other businesses. Total global damage exceeded $10 billion. This matters because unlike kinetic weapons, cyberweapons cannot be geographically contained once deployed. A missile hits a specific coordinate; a worm propagates through any vulnerable network it can reach. The interconnected nature of global IT infrastructure means that a weapon designed to disrupt one adversary's systems will inevitably find pathways into allied, neutral, and civilian networks. The Stuxnet worm, designed by the US and Israel to sabotage Iranian centrifuges, spread to over 100,000 computers in 115 countries. The strategic consequence is that offensive cyber operations carry an inherent risk of escalation and blowback that policymakers consistently underestimate. When a cyberweapon designed to target an adversary's military network instead takes down hospitals, power grids, or financial systems in third-party countries, it can trigger diplomatic crises, economic disruption, and even unintended military escalation if the affected nation misattributes the attack. This persists because the intelligence and military agencies that develop cyberweapons operate under classification regimes that prevent meaningful oversight of deployment risk assessments. There is no equivalent of environmental impact review for cyberweapons. The NSA's Tailored Access Operations and equivalents in other nations stockpile zero-day exploits and deploy them with minimal analysis of second-order propagation effects. The Shadow Brokers' 2017 leak of NSA tools — which enabled WannaCry and contributed to NotPetya — demonstrated that even the storage of these weapons poses existential risks to civilian infrastructure.

Military AI systems designed to identify enemy combatants and military targets from sensor data — satellite imagery, drone feeds, signals intelligence — produce false positives at rates that would be considered catastrophic in any other domain. Israel's AI-assisted targeting system known as 'Lavender,' as reported by +972 Magazine and Local Call in 2024, reportedly generated a database of 37,000 suspected militants in Gaza, with an acknowledged error rate that military sources described as roughly 10%. That means potentially thousands of civilians were flagged as legitimate military targets by an algorithm. The immediate consequence is that commanders who rely on these systems to accelerate targeting decisions are making life-or-death calls based on probabilistic outputs they may not fully understand. When a system flags an individual as a combatant with 90% confidence, the missing 10% represents real human beings. At scale — when the system is processing tens of thousands of targets under time pressure — even a small error rate translates to massive civilian harm. This cascades into a strategic problem: civilian casualties generated by AI targeting erode the legitimacy of military operations, fuel insurgent recruitment, and damage alliances. The US military's own studies after Kabul drone strikes in 2021 — where an aid worker and seven children were killed based on pattern-of-life analysis — showed how algorithmic confidence can produce catastrophic errors that undermine the very mission the technology is meant to support. The problem persists because there is no standardized testing or certification regime for military AI targeting systems. Unlike aviation or medical devices, there is no independent body that audits these systems' accuracy against representative datasets before deployment. Militaries develop and evaluate their own systems internally, with classification preventing external scrutiny. The pressure to accelerate kill chains — reducing the sensor-to-shooter loop from hours to minutes — creates institutional incentives to accept higher error rates in exchange for speed.