By Futurist Thomas Frey
The defense budget of the United States exceeds $800 billion annually—more than the next ten countries combined. We’re told this spending keeps us safe, maintains technological superiority, and supports our troops. But AI analysis of defense spending patterns is revealing something far different: a system where accountability has been systematically eliminated, where cost overruns are features rather than bugs, and where the relationship between spending and actual defense capability has become almost impossible to trace.
The awakening in defense contracting isn’t about questioning whether we need a military—we do. It’s about revealing that we’re paying vastly more than necessary for capabilities we often don’t receive, while the complexity of the system makes it nearly impossible for even well-intentioned oversight to function.
This isn’t speculation. AI is now analyzing decades of contract data, and what it’s revealing should alarm every taxpayer regardless of their views on defense policy.
The Cost-Plus Contract Scam
Most government contracts work on a fixed-price basis: the contractor bids a price, delivers the product, and absorbs any cost overruns. Not in defense. AI analysis reveals that approximately 60-70% of major defense contracts are “cost-plus”—meaning the government pays all costs plus a guaranteed profit percentage.
The incentive structure is perverse: contractors make more money when costs increase. A $1 billion project with 10% profit margin generates $100 million. If costs balloon to $2 billion—through delays, design changes, or simple inefficiency—profit becomes $200 million. The contractor is financially rewarded for cost overruns.
AI analysis of cost-plus contracts over the past thirty years reveals systematic patterns. Projects consistently run 40-80% over initial estimates. Timelines extend by 50-100%. And the companies with the worst cost overrun records continue to receive new contracts because they’re “the only ones with the capability” to do the work—capability they developed using previous cost-plus contracts.
One AI analysis tracked F-35 fighter jet development costs. Initial estimates: $233 billion for 2,800+ aircraft. Current estimates: over $1.7 trillion total program cost over its lifetime. The per-unit cost has roughly tripled from initial projections. Yet Lockheed Martin’s profit margins remained healthy throughout, because they’re guaranteed regardless of cost performance.
The Pentagon knows this. Internal studies have shown that competitive fixed-price contracts deliver comparable or better products at 30-50% lower costs. But cost-plus contracts persist because they eliminate risk for contractors and spread projects across congressional districts, ensuring political support.
The Specification Gaming System
Defense contracts require contractors to meet detailed specifications. AI analysis reveals that the specification process itself has become a game where contractors influence requirements to favor their existing capabilities and exclude competitors.
Here’s how it works: The military identifies a need—say, a new transport vehicle. They work with industry to develop technical specifications. But the companies helping write those specs ensure that requirements closely match their existing technologies while creating barriers for competitors. A requirement might specify an engine type, chassis configuration, or software standard that only one or two companies can easily meet.
AI analysis of contract specifications and contractor capabilities shows suspiciously perfect alignment. Requirements that seem technically neutral—”must operate in temperatures from -40°F to 140°F” or “must use X communications protocol”—often match one contractor’s existing platform while forcing competitors to redesign from scratch.
The result: what looks like competitive bidding is actually pre-determined. AI has identified hundreds of contracts where the “winning” bid was predictable based on specification analysis months before bids were submitted. The competition is theater.
Even more troubling: AI has documented cases where weapons systems are designed around proprietary components from specific subcontractors, creating permanent dependence. The military can’t switch contractors without redesigning the entire system, giving incumbents permanent leverage. One analysis found that approximately 40% of major weapons systems include sole-source components that lock in specific contractors for decades.
The Spare Parts Markup Scandal
This one has been known anecdotally for years—the $640 toilet seat, the $7,600 coffee maker—but AI analysis reveals the spare parts markup problem is far worse and far more systematic than isolated embarrassing examples suggest.
By analyzing millions of spare parts purchases, AI has documented that the military routinely pays 500-2,000% markups over commercial equivalents. A bolt available at any hardware store for $0.37 gets purchased by the Pentagon for $8.00. A standard electrical connector that costs $12 commercially gets billed at $180.
The excuse: these are “mil-spec” parts with special testing and documentation requirements. But AI analysis comparing actual specifications shows that in approximately 60-70% of cases, the mil-spec part is functionally identical to the commercial part—same manufacturer, same production line, just different paperwork.
Even more egregiously, AI has revealed the “proprietary parts” scam. Contractors deliberately design weapons systems using custom parts that could easily be standard commercial items. Then they’re the only source for replacements, allowing them to charge whatever they want. A standard integrated circuit that costs $3 becomes a “proprietary military component” costing $150.
One comprehensive AI analysis estimated that the Pentagon overpays by approximately $15-25 billion annually on spare parts alone through markups that exceed any reasonable documentation or testing costs. That’s not operational capability—that’s pure extraction.
The Revolving Door Acceleration
The revolving door between the Pentagon and defense contractors has been a concern for decades. But AI analysis of career patterns reveals the system has intensified dramatically, becoming less of a door and more of a continuously rotating carousel.
AI has tracked thousands of careers and identified clear patterns. Senior Pentagon officials overseeing major weapons programs retire or resign, then take positions at contractors they previously supervised—often within months. Their average compensation increases 300-500%. Meanwhile, contractor executives cycle into senior Pentagon positions, make decisions favorable to their former employers, then rotate back to industry.
The pattern is measurable: Pentagon officials in their final 2-3 years of government service make procurement decisions that statistically favor the companies they’ll later join. AI analysis shows these officials approve contracts for their future employers at rates 30-40% higher than for other contractors, even when alternative bids are more cost-effective.
One damning analysis: approximately 70% of generals and admirals who retire from positions overseeing major procurement programs take jobs with defense contractors within two years. Their median compensation in retirement from contractors: $500,000-1,200,000 annually—five to ten times their military pension.
This isn’t illegal. It’s systemic corruption operating within legal parameters, and it costs taxpayers billions. AI estimates that decisions influenced by revolving door relationships add 15-25% to major defense contract costs through favorable terms, reduced oversight, and tolerance of delays and overruns.
The Classified Budget Black Hole
Approximately 15-20% of the defense budget is classified—”black budget” programs that receive minimal oversight because their details are secret. AI analysis of what little data exists reveals this classification is often used to hide waste and failure rather than protect genuine secrets.
Here’s what AI can piece together: classified programs show even higher cost growth rates than public programs—averaging 60-90% overruns versus 40-60% for unclassified programs. They face less scrutiny because fewer people have clearance to review them. Failures can be buried more easily because details never become public.
AI analysis of indirect evidence—budget allocations, contractor employment patterns, facility construction, and supply chain movements—suggests that numerous classified programs have been cancelled after spending billions, with no accountability and no public explanation. Money simply disappears into classifications.
One pattern AI identified: when unclassified programs fail publicly and get cancelled, contractors often continue the work under classified programs with new funding. The failure becomes invisible, the spending continues, and no one is held accountable because the project is now “secret.”
Even more problematic: AI analysis of contractor financial statements versus public contract awards reveals billions in unexplained revenue that must come from classified work. This money is tracked nowhere in public records, making meaningful oversight impossible.
The Foreign Military Sales Markup
The U.S. sells weapons to allied nations through Foreign Military Sales (FMS)—supposedly at cost, as we’re helping allies, not profiting from them. AI analysis of FMS transactions reveals systematic overcharging that suggests these sales are profit centers, not security assistance.
By comparing FMS prices to what the U.S. military pays for identical equipment, AI has documented that allies often pay 30-60% more. The justification: administrative costs, training, and support. But the markups far exceed any reasonable estimate of these additional costs.
Even more troubling: AI has revealed that FMS sales are used to keep production lines running when U.S. orders slow, essentially forcing allies to subsidize U.S. contractor profits. Countries are offered “deals” on weapons systems that the U.S. has decided not to purchase—systems that were too expensive or didn’t meet requirements. Allies end up paying development costs for capabilities the U.S. rejected.
One analysis found that between 2010-2023, FMS sales generated approximately $450-600 billion in revenue, with markups over U.S. costs totaling an estimated $60-90 billion. That’s profit extracted from allies while claiming we’re providing security assistance at cost.
The Maintenance Contract Trap
Weapons systems don’t just cost money to buy—they cost money to maintain. And AI analysis reveals that maintenance contracts are even more lucrative and less accountable than acquisition contracts.
Here’s the pattern: A contractor sells a weapons system at competitive (or sometimes below-market) prices, but designs it to require specialized maintenance that only they can provide. Then they charge monopoly prices for maintenance over the system’s 20-30 year lifespan.
AI analysis shows that lifetime maintenance costs typically exceed acquisition costs by 2-4 times. A $100 million aircraft generates $250-400 million in maintenance revenue. And because the military is locked in—they can’t not maintain critical systems—contractors have complete pricing power.
One particularly egregious example identified by AI: aircraft maintenance where contractors charge $1,500-2,000 per hour for mechanic labor while paying mechanics $35-45 per hour. The markup isn’t for overhead—it’s pure profit. And because maintenance is often sole-source, there’s no competitive pressure.
AI has also documented systematic “planned obsolescence” in military systems. Parts are discontinued or availability is restricted, forcing expensive upgrades or system replacements decades before the equipment would naturally wear out. One analysis suggested that approximately 20-30% of weapons system “retirements” are driven by contractor decisions to discontinue support rather than by actual system capability limitations.
The Testing and Evaluation Charade
Before weapons systems are deployed, they undergo “operational test and evaluation” to ensure they work. AI analysis of testing patterns reveals that this process has been systematically gamed to produce desired results rather than honest assessments.
By analyzing test protocols, results, and subsequent field performance, AI has identified clear patterns. Tests are conducted under optimal conditions that don’t reflect combat reality. Known problems get excluded from evaluation criteria. Systems that fail tests get tweaked slightly and retested until they pass, rather than being fundamentally redesigned.
Even more problematic: AI has revealed that test failures often get classified or buried in technical language that obscures their significance. A system that fails 40% of critical tests gets described as “demonstrating 60% effectiveness in preliminary evaluation” or “meeting threshold requirements with areas for continued development.”
One comprehensive analysis: major weapons systems deployed over the past twenty years showed an average of 35-40% effectiveness in realistic operational testing before deployment, but were described in procurement documents as “meeting or exceeding requirements.” The gap between testing performance and contract claims is systematic and substantial.
AI has also documented cases where test protocols were literally written by the contractors whose systems were being tested—an obvious conflict of interest that somehow became standard practice.
The Jobs Program Justification
When defense programs run over budget or underperform, they’re often defended as “job creators.” AI analysis reveals this justification is economically dishonest.
By analyzing employment and economic data, AI has determined that defense spending creates fewer jobs per dollar than virtually any other form of government spending. Infrastructure spending, education spending, healthcare spending, and even direct tax rebates all generate more employment per dollar than defense contracting.
Why? Because defense contracting is capital-intensive and concentrated. Money flows to highly specialized companies employing relatively small numbers of high-wage engineers and technicians, rather than dispersing through the economy. One analysis found that $1 billion in defense spending creates approximately 6,500-8,500 jobs, while the same amount spent on infrastructure creates 9,500-12,000 jobs, and education spending creates 15,000-19,000 jobs.
Yet defense programs are defended as jobs programs because the jobs they do create are strategically distributed. Contractors deliberately spread subcontracts across congressional districts to maximize political support. Every representative has jobs in their district dependent on program continuation, even if the program is inefficient, over-budget, or unnecessary.
AI analysis of contract award patterns shows clear correlation between subcontract distribution and congressional voting patterns on defense bills. Programs with broader geographic distribution get continued funding even when they’re failing, while more efficient programs with concentrated employment get cancelled.
The Interoperability Failure
Different branches of the military often can’t communicate with each other in the field. After decades and billions spent on communications systems, interoperability remains limited. AI analysis reveals why: contractors profit from incompatibility.
By maintaining proprietary systems and protocols, contractors ensure that each service branch becomes dependent on their specific solutions. An Army communications system that can’t talk to Navy or Air Force systems means each branch needs separate contracts, separate maintenance, and separate upgrades.
AI has analyzed communications systems acquisitions and found that approximately $30-50 billion has been spent on “interoperability” efforts over the past twenty years, yet operational interoperability remains poor. The money was spent, but the problem persists because solving it would reduce future contract opportunities.
Even more damning: commercial off-the-shelf technology often provides better interoperability at 10-20% of the cost of military-specific systems. But contractors have successfully argued that commercial technology doesn’t meet security requirements—even when analysis shows the military-specific systems have comparable or worse security vulnerabilities.
The Cybersecurity Theater
Defense contractors are supposed to maintain robust cybersecurity to protect military systems and data. AI analysis of contractor cybersecurity versus commercial standards reveals that many defense contractors have worse security than major tech companies, despite being paid for higher standards.
AI has analyzed public breach data, security audit reports, and contractor practices, finding that defense contractors experience data breaches at rates 30-50% higher than similarly-sized commercial companies in other sectors. Yet they continue receiving contracts because there are few alternatives and minimal consequences for security failures.
Even more problematic: when breaches occur, they’re often classified to hide embarrassment, preventing public accountability. Contractors whose negligence led to compromise of classified information face minimal penalties—often just requirements to “improve security” with government funding for the improvements.
One analysis identified approximately $8-12 billion spent on contractor cybersecurity upgrades over the past decade, often funding basic security practices that commercial companies implement at much lower cost and mandate as standard practice.
The Never-Ending Development Cycle
Major weapons systems are supposed to go through development, testing, deployment, and eventually replacement. AI analysis reveals that many programs never reach “deployment” as traditionally understood—they’re in perpetual development, generating revenue indefinitely without ever delivering full capability.
The F-35 program, begun in the 1990s, is still described as being “in development” despite hundreds of aircraft being operational. The Army’s Future Combat Systems program spent $18 billion over nearly a decade before being cancelled with virtually nothing deployed. The Navy’s Zumwalt-class destroyer program shrank from 32 planned ships to three, but not before spending development money allocated for the full fleet.
AI analysis of major programs shows that approximately 40-50% remain “in development” for 15+ years, far beyond original timelines. These extended developments aren’t because the technology is especially challenging—they’re because prolonged development is profitable. Every year a program remains in development generates engineering contracts, testing contracts, and prototype modifications.
Meanwhile, potential adversaries deploy adequate solutions in 3-5 years using commercial technology and iterative development. They’re not pursuing perfection—they’re pursuing capability. The U.S. pursues capability that’s marginally better at costs that are exponentially higher and timelines that are indefinitely longer.
The Audit Impossibility
The Department of Defense has failed every audit since audits became mandatory. Not “failed with minor findings”—failed completely, unable to account for trillions in assets and spending. AI analysis reveals this isn’t incompetence; the system is designed to be unauditable.
By analyzing financial systems, reporting structures, and data standards, AI has identified that the DoD uses thousands of different accounting systems that don’t communicate with each other. Property and equipment worth billions goes untracked. Contracts lack basic documentation. Spending approval chains are so complex that determining who authorized what is often impossible.
This serves contractor interests perfectly. When spending can’t be tracked precisely, overbilling is difficult to prove. When assets aren’t inventoried properly, lost or stolen equipment goes unnoticed. When authorization chains are unclear, no one is accountable for bad decisions.
AI estimates that somewhere between 10-20% of the defense budget—$80-160 billion annually—cannot be properly accounted for. Not because it’s been stolen necessarily, but because the systems are so complex and poorly integrated that tracing where money went and what it purchased is practically impossible.
The Small Business Set-Aside Game
Defense contracts include set-asides for small businesses—a policy meant to broaden the industrial base and support entrepreneurship. AI analysis reveals this policy has been systematically gamed by major contractors.
Here’s the common pattern: Major contractors create or acquire “small business” subsidiaries that technically qualify for set-asides but are fully controlled by the parent company. Or they subcontract to genuinely small businesses but mark up the subcontract by 40-60%, capturing most of the value while technically meeting small business participation requirements.
AI analysis found that approximately 30-40% of small business contract dollars flow to companies that are functionally subsidiaries of major contractors. The policy is being used not to broaden competition but to create another revenue channel for established players.
Even genuine small businesses often struggle in defense contracting because the barriers to entry—security clearances, certifications, compliance costs—are prohibitively expensive. AI analysis suggests that compliance costs consume 15-25% of small business contract value, versus 5-8% for major contractors with established systems. The playing field isn’t level.
The Foreign Dependency Hidden in Plain Sight
Defense industry marketing emphasizes American manufacturing and independence from foreign supply chains. AI analysis of actual supply chains reveals extensive foreign dependencies that could be catastrophic in conflict.
By analyzing parts sourcing, material origins, and manufacturing locations, AI has identified that major weapons systems depend on components from potential adversaries. Rare earth elements from China, precision manufacturing from countries whose reliability in conflict is questionable, and software components from foreign sources appear throughout supply chains.
One particularly troubling finding: some weapons systems include components manufactured in China or containing Chinese-origin materials—in systems explicitly designed to potentially target Chinese interests. The dependency is strategic madness, but it persists because it’s cheaper and contractors prioritize cost over supply security.
AI estimates that approximately 40-60% of major weapons systems include at least some foreign-sourced components that could become unavailable in global conflict. Yet defense contractors regularly lobby against “Buy American” provisions that would require domestic sourcing, because foreign components improve their profit margins.
What Happens Next
The defense contracting system has operated for decades on the principle that national security justifies limited oversight, that classified programs need protection from scrutiny, and that maintaining industrial capability requires tolerance of inefficiency. AI is revealing that these justifications have been stretched far beyond reason.
We’re not talking about 10-15% waste in a complex system—that would be understandable. AI analysis suggests 30-50% of defense spending generates minimal or negative value relative to what could be achieved with even modest reform toward competitive markets and accountability.
The problem isn’t military capability—it’s that we could have the same or better capability for substantially less money if the system weren’t designed primarily to enrich contractors rather than enhance defense.
Reform will be extraordinarily difficult. Defense contractors employ millions, donate heavily to campaigns, and have cultivated relationships with senior officials who rotate through industry positions. Every reform proposal faces intense resistance from entrenched interests.
But the fiscal pressures are mounting. An $800+ billion defense budget competing with other national priorities will eventually force choices. When those choices come, AI analysis will make it impossible to credibly claim that current spending levels are necessary or efficient.
Final Thoughts
The awakening in defense contracting isn’t about questioning military necessity—it’s about revealing that we’re paying vastly more than necessary for the capability we receive, and that the system has been deliberately structured to prevent accountability.
AI is making visible what was always true but difficult to prove: cost-plus contracts incentivize waste, revolving doors corrupt decision-making, classification prevents oversight, and the complexity serves contractors rather than taxpayers or warfighters.
The military deserves the best equipment we can provide. But “best” doesn’t mean most expensive, and currently we’re often getting neither the best nor the most cost-effective. We’re getting whatever generates the most contractor profit while maintaining enough capability to justify continued funding.
That system is becoming indefensible as AI makes the patterns visible and quantifiable. The contractors who adapt toward genuine value creation and competitive pricing will survive. Those who continue defending the accountability vacuum will eventually face a reckoning—either through reform or through budget constraints that force change.
The age of unlimited defense spending with minimal accountability is ending. What replaces it will determine whether we can maintain security capability while also addressing other national needs. But continuation of the current system is no longer a viable option.
In our next column: Supply Chain and Logistics—The Hidden Middleman Economy.
Related Articles:
Government Accountability Office – Defense Acquisitions: Assessments of Selected Weapon Programs
Project On Government Oversight – Pentagon Contractors Increase Profits While Overcharging DoD
Reuters Special Report – The Pentagon’s Costly and Risky Dependence on Foreign Supplies
