Is defense tech becoming a bubble?
In our defense tech market deck, you will find everything you need to understand the market
SUMMARY
Yes, defense tech is becoming a selective bubble. Military demand is real and still growing, but late-stage private valuations and crowded startup categories have moved well ahead of contracts, production capacity, and credible exit paths.
The underlying customer market is not speculative. Global military spending reached nearly $2.9 trillion in 2025, European equipment budgets rose sharply, and the largest contractors entered 2026 with years of work already sitting in their backlogs.
The excess is concentrated rather than universal. A propulsion supplier with repeat orders and several platforms is not carrying the same risk as a drone or autonomous-ship startup valued on factories and production volumes that barely exist yet.
The funding boom is also narrower than the headline numbers suggest. Three transactions supplied more than half of the capital raised in the first half of 2026, so the market is being pulled upward by a few giant checks rather than a sudden improvement across every defense startup.
Investors have already priced in a major transfer of procurement from legacy contractors to startups. That transfer has only started: even the largest defense tech companies still receive less than 1% of Pentagon contract spending.
The valuation problem is not simply that the multiples are high. Investors are applying software-style growth expectations to businesses that must finance factories, inventory, testing, secure supply chains, and long military qualification cycles.
Government contract headlines can make progress look faster than it is. Prototype awards, framework agreements, and contract ceilings improve a startup’s position, but they are not the same as funded production orders or guaranteed backlog.
Manufacturing is now the decisive test. Several companies have raised enough money to build enormous plants before investors can see whether those plants will deliver thousands of reliable units on schedule and at the promised cost.
Basic drone assembly may become one of the largest markets and one of the worst businesses. As expendable airframes become easier to copy, margins are likely to migrate toward autonomy, navigation, communications, propulsion, sensors, and integration.
Defense AI can still justify premium economics, but only when the same software works across many programs without turning each deployment into a custom engineering project. Palantir proves the model can work; it does not prove every company labelled “defense AI” has built it.
A correction would probably leave military demand intact. The likely outcome is harsher funding concentration, failed factory plans, acquisitions at disappointing prices, and a small group of survivors growing into genuinely large defense companies.
This market map, featured in our defense tech market deck, highlights top companies and startups in the defense tech market
What would a defense tech bubble actually look like?
A defense tech bubble would mean investors are valuing companies far above the revenue and cash flow that governments can realistically generate for them.
The industry does not need to be fake for a bubble to form. Governments may genuinely need millions of drones, better air defense, autonomous aircraft, and faster battlefield software. Investors can still pay too much for the companies supplying them.
We should also avoid treating the whole sector as one market. A profitable propulsion supplier with several customers carries very different risks from a drone startup valued on a future factory that has barely started production. The same applies to reusable software and custom engineering work, even when both are sold as “defense AI.”
The useful test is straightforward. Are government orders, production volumes, and repeat purchases catching up with the amount of capital invested and the valuations assigned to private companies?
For a small group of companies, yes. Across the wider sector, the gap is still growing.
If you want more recent data on this point, please see our latest defense tech market report.
Why is defense tech suddenly attracting so much money?
Defense tech has become fashionable because it combines rearmament, artificial intelligence, and advanced manufacturing in one investment story.
Several barriers that kept venture capital away have also disappeared. Working on weapons is more socially accepted in Silicon Valley. Governments are openly asking startups to challenge traditional contractors. Large institutional investors now see national security as a long-term investment theme rather than a specialist niche.
The wars in Ukraine and the Middle East gave the pitch a visible technological foundation. Cheap drones have destroyed equipment worth millions of dollars. Software has shortened targeting cycles. Electronic warfare has become central to daily combat. Western governments have also discovered that their factories cannot replenish missiles, shells, and interceptors quickly enough.
Money arrived fast. According to The Wall Street Journal, defense tech startups raised $16.8 billion during the first half of 2026. Crunchbase had already counted $14.6 billion in the first five months, more than the $9.6 billion it recorded for the whole of 2025.
The investor base changed too. Recent rounds for Anduril, Helsing, Shield AI, and Quantum Systems included pension money, crossover funds, banks, private-equity firms, and established aerospace companies. These investors can write much larger checks than the specialist funds that built the first generation of defense startups.
Defense tech was probably underfunded several years ago. Now the larger danger is that capital is moving faster than governments can turn new technology into recurring production orders.
As this chart shows, and as featured in our defense tech market deck, search interest in defense tech has risen sharply
Is the underlying military demand really this strong?
Yes. Military demand is strong enough to support years of defense-industry growth.
SIPRI estimates that global military expenditure reached $2.887 trillion in 2025. Spending increased for the eleventh consecutive year and stood 41% above its level a decade earlier after adjusting for inflation.
Growth has slowed, which deserves attention. Global spending rose 2.9% in real terms in 2025, compared with 9.7% the previous year. The published total was still roughly $169 billion above the prior year, although inflation and later revisions complicate that comparison.
Europe remains much hotter than the global average. NATO says its European members and Canada spent $574 billion in 2025, almost 20% more in real terms than the year before. The European Defence Agency separately calculated that the EU’s 27 members spent €418 billion, up 20%.
The money is reaching equipment. EU equipment procurement climbed 26% in real terms to €115 billion. Lockheed Martin ended 2025 with a record $194 billion backlog against $75 billion of annual sales. BAE Systems reported £83.6 billion of backlog against £30.7 billion of sales. Rheinmetall’s backlog rose from €46.9 billion to €63.8 billion, although its figure includes framework agreements as well as binding orders.
These order books cover several years of production. Governments are already buying, rather than merely publishing spending plans.
| Demand measure | Latest figure | What it tells us |
|---|---|---|
| Global military spending | $2.887 trillion | The customer market is still expanding |
| Growth over ten years | 41% in real terms | Rearmament predates the latest funding rush |
| NATO Europe and Canada | $574 billion | European demand is rising especially quickly |
| EU equipment procurement | €115 billion | A meaningful share is reaching weapons and hardware |
| Lockheed Martin backlog | $194 billion | Established programs have years of visible demand |
| BAE Systems backlog | £83.6 billion | The spending cycle extends beyond one conflict |
Is startup funding outrunning defense procurement?
Yes. Startup funding is currently moving much faster than the share of defense procurement won by startups.
The figures are not directly comparable: investors finance factories, research, and several years of growth, while contract data records purchases during a specific period. Even so, the distance between them is striking.
The Wall Street Journal found that Pentagon spending on the 15 largest defense tech startups had tripled since 2022. Those companies still received less than 1% of total Pentagon contract spending.
The broader NatSec100 group captured about 0.5% of Department of Defense contract obligations in the latest available fiscal year. The Pentagon’s main suppliers remain Lockheed Martin, RTX, General Dynamics, Boeing, Northrop Grumman, and other established contractors.
Meanwhile, startups raised $16.8 billion in six months. That is more than four times the roughly $4 billion the Pentagon reportedly awarded to hundreds of defense tech startups during the previous fiscal year.
Dividing those figures would produce a meaningless valuation ratio. The useful point is simpler: investors have already funded a major redistribution of defense spending while the redistribution itself has barely started.
It can still work. Startup contract revenue could grow several times over without taking a large share of the Pentagon budget. Current prices, though, require that growth to arrive quickly and concentrate around today’s funded companies.
This chart, included in our defense tech market deck, shows annual VC investment in defense tech startups
Have private defense tech valuations gone too far?
Yes. Several leading defense tech valuations now require near-perfect growth for many years.
Anduril’s latest $5 billion round valued the company at $61 billion. Barron’s estimates that this equals roughly 14 times expected 2026 sales, compared with around two to three times sales for established contractors such as Northrop Grumman and L3Harris.
Anduril can justify a premium. Revenue is reportedly growing quickly, it has won major programs, and it operates across software, drones, missiles, sensors, and underwater systems. A 14-times multiple still leaves little room for delayed programs, weaker margins, or an expensive factory ramp-up.
Helsing looks more aggressive. Its latest $1.8 billion round valued the company at $18 billion. The Financial Times compared that price with projected 2026 revenue of €441 million and calculated a multiple of roughly 32 times forward sales.
Helsing has real contracts, including German orders for strike drones and an AI combat cloud. Two contracts worth a combined €491 million do not support an $18 billion valuation by themselves. Investors are paying for Helsing to become a leading European defense platform across several countries and product categories.
Shield AI’s $12.7 billion valuation works out at roughly 21 times projected sales using the estimates cited by the Financial Times. Quantum Systems looks less stretched at around 8.5 times. Saronic reached a $9.25 billion valuation while still building the shipyard meant to support its future scale.
Private revenue figures are estimates rather than audited public disclosures, so these multiples are approximate. The broad conclusion is still hard to avoid: investors are applying high-growth technology valuations to companies that also face aerospace testing, factory construction, inventory, and supply-chain costs.
| Company | Latest valuation | Estimated forward sales multiple | What investors are assuming |
|---|---|---|---|
| Anduril | $61 billion | About 14× | Rapid growth across several major programs |
| Helsing | $18 billion | About 32× | European leadership in AI and autonomous weapons |
| Shield AI | $12.7 billion | About 21× | Hivemind becomes a reusable autonomy standard |
| Quantum Systems | About $8 billion | About 8.5× | Profitable expansion across multiple domains |
| Legacy contractors | Varies | Commonly 2–3× | Slower growth with proven production and cash flow |
If you want more recent data on this point, please see our latest defense tech market report.
Are a few mega-rounds making the whole market look hotter?
Yes. A handful of enormous rounds are creating much of the apparent defense tech funding explosion.
Anduril raised $5 billion, Shield AI arranged $2 billion of equity and preferred financing, and Saronic raised $1.75 billion. Together, those three transactions accounted for $8.75 billion, or approximately 52% of the $16.8 billion invested during the first half of 2026.
Quantum Systems then raised $1.2 billion and Helsing added $1.8 billion. Across less than four months, five companies announced a combined $11.75 billion of financing.
Deal counts have increased much more slowly. Crunchbase counted 107 announced defense tech rounds through the first five months of 2026, a pace only slightly ahead of the 206 rounds completed during all of 2025. Dollars exploded because the checks became much larger, not because funding suddenly reached every defense startup.
That creates two different markets. A small group of companies can raise enough capital to buy suppliers, build factories, and survive long procurement cycles. Smaller companies still face the same slow customers, expensive hardware, and limited bridge financing.
This concentration may prove rational. Defense procurement naturally favors a few trusted suppliers that can pass security reviews, integrate with existing systems, and support products for decades.
The trouble starts when investors assume dozens of startups can all become the next Anduril. Governments may want more competition, but they rarely sustain large numbers of suppliers for the same military capability.
This chart, included in our defense tech market deck, shows why Anduril is winning in defense tech
Are governments buying startup weapons or mostly running pilots?
Governments are buying more defense technology from startups, although production contracts remain concentrated among a few companies.
The progress is real. The U.S. Army awarded an Anduril-led team $99.6 million to test its Next Generation Command and Control system. The program has since moved from early prototyping toward division-level delivery. Shield AI received a production contract to provide Hivemind mission-autonomy software for the Air Force’s Collaborative Combat Aircraft program.
A more recent agreement could take Anduril’s Barracuda-500 missile into production at thousands of units per year, with initial deliveries planned for 2027. That is the kind of repeatable, high-volume program that can support a large defense company.
The wider conversion rate remains weak. The Government Accountability Office found that the Defense Innovation Unit awarded 450 prototype agreements between fiscal years 2016 and 2023. Only 62 projects had reached a production agreement or conventional contract.
DIU reports a 51% transition rate when it counts only projects with completed prototypes. GAO’s broader calculation puts the 62 transitions at roughly 14% of all prototype awards. Both figures are valid, but a startup deciding whether a prototype award will eventually support the business cannot ignore the broader denominator.
Large contract announcements can also mislead readers. The Army’s enterprise agreement with Anduril carries a ceiling of up to $20 billion over ten years and consolidates around 120 existing and future procurement actions. The Army has also said that future requirements will remain competitive.
That agreement gives Anduril a much easier route to future orders. It does not place $20 billion of guaranteed revenue in the company’s backlog.
The Pentagon is now taking startups seriously. It still selects them one program, task order, and production decision at a time.
Can defense startups manufacture at the scale their valuations require?
Manufacturing has become the hardest test for highly valued defense startups.
The leading companies have moved far beyond the normal software startup model. Anduril is developing Arsenal-1, a manufacturing campus planned to exceed five million square feet. Saronic is pursuing a shipyard project expected to cost roughly $3.2 billion. Shield AI is financing a new aircraft program while integrating Aechelon, the simulation company it recently acquired.
Factories change the risk profile. They require land, equipment, inventory, technicians, secure suppliers, and quality-control systems before revenue reaches full scale. A delayed software launch may cost a company several months. A delayed missile factory can consume hundreds of millions of dollars while customers wait.
Current supply chains make the challenge worse. The Government Accountability Office says more than 200,000 suppliers support the U.S. defense industrial base, yet the government has limited visibility into where many lower-tier components and materials originate. Motors, electronics, propellants, batteries, and specialist machine tools can all become bottlenecks.
Anduril has shown some encouraging progress. It says it delivered more than one normal month of Pulsar electronic-warfare production in a single week during an urgent deployment. Its autonomous underwater vehicles have also accumulated thousands of operating hours. At least part of the company has moved beyond prototypes.
We have much less evidence for several other highly valued startups. A planned factory, production target, or government memorandum tells us what a company hopes to build. Units delivered on time, at the promised cost, and with acceptable failure rates tell us whether the industrial model works.
Investors are funding the factory phase before most of those results are visible. A large share of the sector’s financial risk now sits there.
This chart, included in our defense tech market deck, shows annual funding in defense tech startups
Are military drones becoming a commodity business?
Military drones will remain a huge market, but basic drone manufacturing is already becoming a commodity business.
Ukraine produced more than four million drones in 2025 and planned to exceed seven million in 2026, according to its Defense Ministry. Volumes at that scale permanently change how militaries think about reconnaissance, targeting, and strike capacity.
They also change the economics for suppliers. A drone expected to survive for years can support high prices, long maintenance contracts, and slow upgrade cycles. A drone expected to survive one mission is bought more like ammunition.
FPV airframes are relatively easy to copy. Components can be sourced from several manufacturers, designs change quickly, and users care more about current battlefield performance than the brand printed on the aircraft. A successful model can lose its usefulness within months when electronic warfare, detection systems, or operating tactics change.
The better businesses sit deeper in the stack. Navigation without GPS, jam-resistant communications, electronic-warfare software, sensors, propulsion, target recognition, and counter-drone integration are harder to reproduce than a basic airframe.
Secure supply chains may also command a premium. Western militaries want to reduce their dependence on Chinese motors, cameras, radios, and flight controllers. A domestically produced component can remain valuable even when the complete drone becomes cheaper.
The military drone market will grow while many drone manufacturers disappear. High demand can support huge unit volumes, but the margins will move toward scarce components, autonomy, and integration.
If you want more recent data on this point, please see our latest defense tech market report.
Can defense AI really earn software margins?
Sometimes. Defense AI can earn software margins when the same product works across many programs without becoming a new engineering project each time.
Palantir shows what that model can look like. The company generated $4.475 billion of revenue in 2025, up 56%. It reported a 32% operating margin and adjusted free cash flow of $2.27 billion, equal to 51% of revenue.
Those economics came after more than twenty years of product development, customer work, and government contracting. Palantir now has software platforms that can be reused across military agencies and commercial customers.
Shield AI is trying to build a similar platform around autonomy. Its Hivemind software has reportedly operated across 30 classes of vehicles. The Air Force’s decision to treat mission autonomy as a separate capability in the Collaborative Combat Aircraft program gives Shield AI a chance to sell its software across different aircraft rather than tying it permanently to one airframe.
The outcome is still uncertain. Military AI often needs classified data, custom sensors, specialist testing, and integration with old systems. Engineers may have to remain involved throughout every deployment. Costs then rise with revenue, producing something closer to systems integration than a scalable software platform.
The term “defense AI” tells us very little about the economics. We need to know how often the code can be reused, how much customization each customer requires, and whether the company can charge separately for software.
Companies that solve those problems may deserve high multiples. The rest are hardware or engineering businesses carrying a fashionable label.
This chart, included in our defense tech market deck, compares the main business model options for defense AI contractors
Can Europe turn bigger defense budgets into startup revenue?
Europe has enough defense money to build major startups, but its fragmented procurement system still slows them down.
The headline figure is impressive. EU defense spending reached €418 billion in 2025, and the European Defence Agency projects €454 billion in 2026. Nineteen countries have also applied for the EU’s €150 billion SAFE loan program, which is intended to support joint defense procurement.
The addressable market becomes smaller as we narrow the figures. Of the €418 billion spent in 2025, €115 billion went toward equipment procurement. Only 24% of equipment spending was collaborative.
That means EU countries jointly procured roughly €28 billion of equipment, equivalent to about 6.6% of total EU defense spending. The rest remained divided across national budgets, personnel, operations, infrastructure, maintenance, and country-specific equipment programs.
A startup selling the same drone or software platform across Europe may therefore need different contracts, requirements, industrial partners, and security approvals in Germany, France, Italy, Poland, and Spain. Governments also protect domestic factories and employment.
Europe spends relatively little on research compared with equipment. The EDA calculated €17 billion of defense research and development in 2025, equal to 4% of total expenditure. Most of the recent increase has gone toward closing immediate equipment gaps.
Helsing and Quantum Systems have shown that European companies can raise American-sized rounds. Their challenge now is commercial rather than financial. They need Europe to place larger common orders, standardize requirements, and buy across borders.
Until that happens, Europe will have a rapidly growing defense budget without behaving like one unified customer.
Will defense startups actually take share from legacy primes?
Defense startups will take valuable programs from legacy contractors, but the established primes will continue controlling most defense spending.
BCG and Vertical Research estimate that traditional complex systems could still represent more than 80% of the weapons market in 2033. Fighter aircraft, submarines, warships, strategic missiles, radar networks, and large air-defense systems absorb enormous budgets and remain difficult for young companies to build alone.
The startup opportunity is still large. Affordable autonomous systems and software can grow much faster than the wider market while remaining a minority of total spending. Anduril does not need to replace Lockheed Martin to become a very large company.
Recent Army decisions show how the market may develop. Anduril is leading the Next Generation Command and Control implementation for the 4th Infantry Division, while Lockheed Martin is leading a parallel implementation for the 25th Infantry Division. The Army is testing both new and established suppliers rather than handing the entire architecture to one camp.
Incumbents can also buy technology, partner with startups, or use their manufacturing capacity to build startup-designed products. Airbus joined Quantum Systems’ latest funding round. HD Hyundai is working with Anduril on autonomous vessels. L3Harris invested in Shield AI before its latest financing.
Startups will gain influence fastest in autonomy, command software, sensors, low-cost munitions, and uncrewed systems. Incumbents will keep most of the large platforms and long-term sustainment work.
A few startups may eventually join the top tier. Most will end up as suppliers, acquisition targets, or failed challengers.
This chart, featured in our defense tech market deck, shows the share of revenue generated by each customer segment in the defense tech market
Are investors underpricing political risk and weak exit options?
Yes. Investors are paying too little attention to political risk and the limited number of exits available at today’s valuations.
Higher defense budgets give the sector a strong foundation, but they do not protect individual programs. A government can increase military spending while cancelling one drone, changing a missile specification, or moving work toward a domestic supplier.
Budgets also depend on annual appropriations. Continuing resolutions can delay new starts and production increases. Elections can change procurement priorities. Export approvals can remove a foreign market. A change in alliance policy may force companies to rebuild products or supply chains.
Europe adds another layer. National governments often care about local employment, sovereign control, and industrial work share as much as price or technical performance. The best product does not always win the order.
Exit mathematics creates a separate problem. Anduril, Helsing, Shield AI, Saronic, and Quantum Systems have a combined private valuation of roughly $109 billion. Investors need future stock-market listings or acquisitions above those prices before they earn a meaningful return.
Pure defense technology has produced few public companies at that scale. Palantir, AeroVironment, and Kratos provide useful precedents, but they reached public investors with established revenue and years of operating history.
Acquisitions remain the more common exit for defense startups. Those deals often happen early, when an incumbent wants a technology team or access to a program. Early acquisitions can generate good returns from modest valuations. They become harder when the startup is already worth $5 billion, $10 billion, or $20 billion.
S&P Global found that defense tech venture funding reached record levels in 2025 while merger activity slowed. Some industry reports show large exit values by including cybersecurity, nuclear energy, commercial space, and semiconductor companies within defense tech. Those broad categories make the exit market look deeper than it is for autonomous weapons and military hardware.
The sector may produce several major public listings. It now needs them.
If you want more recent data on this point, please see our latest defense tech market report.
Which parts of defense tech look most overheated now?
The most overheated defense tech segments combine crowded competition, thin revenue, and expensive production plans.
Basic drone assembly sits near the top of the list. Governments will buy millions of drones, but airframes are easy to copy and prices are likely to fall. Many manufacturers will struggle to preserve margins once emergency procurement becomes more standardized.
Large autonomous platforms carry a different risk. Ships and combat aircraft could become enormous markets, but valuations already assume factories, military acceptance, and high-volume orders that remain years away.
Defense AI deserves a mixed judgment. A reusable autonomy or command platform can produce excellent economics. Custom software projects with large engineering teams cannot.
Counter-drone systems have urgent demand, although the market is crowded and technical approaches keep changing. Some companies will be stranded when governments settle on fewer sensors, effectors, and command systems.
Propulsion, energetics, navigation, communications, and secure components look more grounded. These areas solve persistent shortages, serve several platforms, and benefit from growth across the sector.
| Segment | Current demand | Bubble risk | Our judgment |
|---|---|---|---|
| Basic small-drone assembly | Very high | Very high | Large volumes, weak differentiation, and falling prices |
| Autonomous ships and aircraft | Rising quickly | High | Valuations depend on factories and orders still being built |
| Defense AI platforms | High | Medium to high | Attractive when software is genuinely reusable |
| Counter-drone systems | Very high | Medium to high | Urgent need, followed by likely supplier consolidation |
| Propulsion and energetics | High | Lower | Scarce capacity supports pricing and repeat demand |
| Secure navigation and communications | Very high | Lower | Needed across many platforms and hard to replace |
| Defense manufacturing technology | High | Lower | Benefits from the wider production bottleneck |
| Traditional major platforms | Strong | Lower | Visible backlogs, mature factories, and proven cash flow |
This chart, included in our defense tech market deck, shows how tactical networking platform technology has evolved over time
Is defense tech becoming a bubble?
Yes. Defense tech is becoming a bubble around private pricing and crowded startup categories, even while military demand keeps rising.
The boom has a solid industrial base. Global military spending is close to $2.9 trillion. European budgets are growing quickly. Equipment procurement is increasing. Legacy contractors have record order books. Governments are placing real orders for autonomy, command software, drones, and lower-cost weapons.
The financial layer has moved much further. Startups raised $16.8 billion in six months while the largest defense tech companies still captured less than 1% of Pentagon contract spending. Three companies absorbed more than half of that funding. Five private companies are now worth roughly $109 billion between them.
Some of those valuations will work. Anduril has built meaningful revenue and won programs that could support a new prime contractor. Palantir has already proved that government software can generate strong growth and margins. Helsing, Shield AI, Saronic, and Quantum Systems have credible products and serious customers.
The weaker companies face a brutal next phase. Pilots must become production orders. Contract ceilings must become funded task orders. Factories must deliver thousands of reliable units. Software must work across several platforms. Investors eventually need public listings or buyers willing to pay more than the latest private round.
Defense spending should remain strong even if private valuations fall sharply. The correction will therefore look different from a market where customer demand disappears. Funding will concentrate further, weaker startups will sell or close, and the survivors will keep growing into a larger defense economy.
The direct answer is clear: defense tech has entered a selective bubble, with the greatest excess in late-stage valuations, basic drone companies, and autonomous platforms priced years ahead of production.
The rearmament boom can last much longer than the current startup valuations.
If you want more recent data on this point, please see our latest defense tech market report.
OUR METHODOLOGY
This analysis tests whether defense tech is becoming a bubble by comparing investor expectations with the evidence that can support them: military spending, equipment procurement, startup contract revenue, repeat production orders, private valuations, manufacturing capacity, software reuse, political exposure, and exit options.
We treated the sector as several different markets rather than one trade. Reusable software, basic drone assembly, propulsion, secure communications, autonomous aircraft, shipbuilding, and traditional major platforms have different margins, capital needs, procurement cycles, and competitive structures.
Military spending and legacy-contractor backlogs were used to establish the strength of underlying demand. Startup funding was then compared with procurement share and contract conversion to see whether capital was moving faster than governments were shifting orders toward newer suppliers.
Private-company sales multiples are approximate because the underlying revenue figures are estimates rather than audited public disclosures. We used them to show the scale of investor expectations, not to present false precision. Contract ceilings, framework agreements, prototype awards, and memoranda were also kept separate from funded production orders and guaranteed backlog.
We looked at mega-round concentration because a few unusually large financings can make the whole market appear hotter than it is. Segment judgments were based on current demand, differentiation, production difficulty, likely pricing pressure, and whether the same technology can be sold across several programs.
We prioritized primary sources and reporting that added specific, checkable figures. Key sources include SIPRI’s Military Expenditure Database, NATO defense-expenditure data, the European Defence Agency’s Defence Data Portal, the U.S. Government Accountability Office’s review of Defense Innovation Unit transitions, U.S. Department of Defense contract announcements, and annual reports from Lockheed Martin, BAE Systems, and Rheinmetall.
Funding, valuation, and company-level claims were checked against Crunchbase Research, the Financial Times, The Wall Street Journal, Barron’s, S&P Global Market Intelligence, company newsrooms, investor-relations disclosures, and official U.S. Army and European Commission materials.
In our defense tech market deck, we identify pain points entrepreneurs should prioritize
Related blog posts
- Is it already too late for defense tech?
- Defense AI: which startup is ahead?
- Military drones: which startup is ahead?
