Cutting-Edge Technologies
NextFin News - The wars in Ukraine and the Middle East are still raging, but in the boardrooms and fund-raising decks of Silicon Valley, defense technology has acquired a new sheen. In the first five months of 2026, venture investors poured more than $14.6 billion into companies in the military, national-security and law-enforcement categories — already eclipsing the sector's previous annual record of $9.6 billion set in all of 2025 — while Anduril Industries closed a $5 billion funding round at a $61 billion valuation and the U.S. Army handed it a contract vehicle worth up to $20 billion. The stardust is unmistakable. The question is whether it is settling on a real industrial base or on a valuation bubble dressed in camouflage.
This is the tension at the heart of the defense-tech moment: private capital is arriving at Cold War intensity, yet federal awards to the very companies being celebrated still amount to "basis points" of the overall defense budget, and only 8 percent of those awards are actual production contracts, up from effectively zero a year ago. The glamour has arrived faster than the factories.
The Numbers Behind the Glow
The headline figures are staggering. Anduril, founded in 2017 by former Palantir engineering director Brian Schimpf alongside Trae Stephens, Palmer Luckey and Matt Grimm, has become the face of the boom. On May 14, 2026, the company closed a $5 billion Series H round led by Thrive Capital and Andreessen Horowitz — the largest defense-tech venture raise on record — at a $61 billion post-money valuation, roughly double the $30.5 billion it commanded just eleven months earlier when Founders Fund led a $2.5 billion round. Schimpf disclosed $2.2 billion of 2025 revenue, also a doubling, and the company now employs about 7,000 people.
The funding arrived alongside contracts. On March 14, 2026, the U.S. Army awarded Anduril an indefinite-delivery, indefinite-quantity vehicle worth up to $20 billion for its AI-driven defense systems — the largest contract the Army has ever issued to a non-traditional contractor. In May, it added a Dutch Ministry of Defence counter-drone contract and an Indo-Pacific missile-defense task order built around its Lattice software. A secondary market mark in March put the company at $95.08 billion, a reminder that the paper value of defense tech is running well ahead of even its frothy primary rounds.
Anduril is not alone. In the ninety days leading into mid-May, four defense-tech financings closed: Shield AI raised $2 billion at a $12.7 billion valuation in March, led by Advent International and JPMorgan Chase, backed by its selection for the U.S. Air Force's Collaborative Combat Aircraft program — the first time mission-autonomy software was procured separately from the aircraft itself; Saronic raised $1.75 billion at a $9.25 billion valuation to mass-produce unmanned surface vessels; and Helsing raised $1.2 billion at $18 billion. Crunchbase data tracked more than $4.5 billion of defense-related venture funding in the first quarter of 2026 alone, a record quarter.
The public markets are participating. Palantir Technologies, the data-analytics and military-software vendor, reported third-quarter 2026 revenue that nearly doubled year over year, with U.S. commercial revenue climbing 148 percent to $984 million — exceeding U.S. government revenue of $953 million for the first time in company history. On November 2, 2026, Palantir raised its full-year 2026 revenue guidance to between $8.530 billion and $8.534 billion, implying roughly 91 percent annual growth, and pushed its adjusted operating-income forecast to between $5.245 billion and $5.249 billion. The Army consolidated dozens of Palantir software contracts into a single enterprise agreement in July 2025 worth up to $10 billion.
The ecosystem's own scorecard confirms the scale. The 2026 NatSec100, produced by the Silicon Valley Defense Group in partnership with J.P. Morgan, counts a cohort that has collectively raised $304.1 billion in lifetime private capital. Three companies — OpenAI, Databricks and Scale AI — account for $235 billion of that. Excluding OpenAI, the cohort raised $118.2 billion, modestly below the 2025 cohort's $131.4 billion, a reflection of SpaceX and Groq graduating from the list rather than any slowdown in underlying activity.
How a Taboo Became a Trend
The speed of the reversal is the story. For years, defense work was a stain on a startup's reputation. In 2018, more than 4,000 Google employees signed a petition demanding the company stop work on Project Maven, a Pentagon program using machine learning to analyze drone surveillance footage; about a dozen engineers resigned, and Google declined to renew the contract. The company's AI principles subsequently vowed that its technology would never be used for military or surveillance purposes. That vow was quietly removed in March 2025.
What changed? Three forces converged. First, the wars in Ukraine and the Middle East turned battlefields into live test beds. Drones and autonomous systems designed in California and Austin could be fielded, iterated and proven in combat within months rather than the decade-long procurement cycles of the legacy primes. "Businesses don't have to pay for pilots, they can just send drones to Ukraine to test the latest firmware," one founder put it.
Second, the economics of software investing broke. Jenny Xiao, a partner at Leonis Capital, argued at a 2024 technology conference that artificial intelligence would drive the cost of building software toward zero, eroding the moat that made SaaS the venture model for two decades. "Software is no longer going to be as good of a business in the next 20 or 30 years," she said. "Mostly because of AI—because AI makes it so easy to build software." Hardware — the ugly duckling of tech, with its factories and supply chains — suddenly looked like the last place where competition could not instantly arbitrage margins away. Marcus Ryu of Battery Ventures called it "a general awakening to the possibility that there is enterprise value creation in other areas besides SaaS software."
Third, the politics shifted. A global turn toward conservatism made defense work respectable again in a valley that had leaned away from it. And the money itself had to be rewired: as Schimpf recalled of the early days, VCs "literally weren't allowed to" invest in defense. "They had to change their LP agreements."
Today the opposite is true. Dedicated defense funds are launching monthly. The 2026 NatSec100 report notes that aggregate private capital investment in defense, as a share of the total defense budget, is now on par with federal spending benchmarks seen during the Cold War — the era in which DARPA bankrolled the research that gave the world the internet and Silicon Valley itself.
The Mechanism: Why This Time the Money Has Teeth
The crucial distinction between this cycle and the defense-tech flirtations of the 2000s is procurement. For most of the post-9/11 era, non-traditional companies could win prototypes and studies but rarely production. The money came from venture capital, not the customer, and the customer's procurement machine was designed to keep them out. That is the structural change now underway.
The 8 percent production-contract share in the 2026 NatSec100 cohort — up from effectively zero — is the single most important number in the report. It means the government is beginning to buy these companies' products at volume, not just fund their experiments. Anduril's $20 billion Army vehicle, Palantir's $10 billion enterprise agreement, Shield AI's autonomy software decoupled from an airframe: these are not pilots. They are the architecture of a repeatable buying relationship.
But the mechanism has a second gear. The wars created urgency, and urgency compresses timelines. Replicator-style pathways, other-transaction authority and modular contracting let the Pentagon buy software the way it buys software — in months, not years. That compression is what allows a company founded in 2017 to stand next to Lockheed Martin and Northrop Grumman in a conversation about the future of American firepower.
"Today, we are living through what is likely a fourth industrial revolution alongside a fourth revolution in military affairs," said Mike Keating, executive director of the Silicon Valley Defense Group. "Our country and national security apparatus cannot afford to lose the global military technology competition for deterrence."
The transmission channel from war to valuation is therefore direct and observable: combat proof creates a procurement anchor; the procurement anchor de-risks the revenue model; de-risked revenue attracts late-stage capital at software multiples; capital funds factories; factories enable production contracts; production contracts justify higher valuations. It is a flywheel, and for the first time it is spinning under its own power rather than being pushed by patriotic venture capital.
The Gap That the Stardust Hides
Here is the uncomfortable counterpoint. The same NatSec100 report that celebrates the boom also warns that "aggregate federal spending on NatSec100 companies still constitutes only basis points of the overall DoD budget," and that "the gap between federal awards and private capital investment continues to widen, creating growing risk for the long-term health of the emerging defense technology ecosystem."
The gap has a number. A funding tracker that followed the 100 best-funded U.S. national-security startups found they had raised roughly $118 billion in venture capital but won only about $16 billion in federal contracts — a ratio of seven dollars of private money for every dollar the government has committed to buy. Anduril is the exception, carrying an Army framework contract worth up to $20 billion.
Consider the arithmetic. The Pentagon's fiscal 2026 budget request is $961 billion, with more than $1 trillion in total national-security spending. The FY2027 request climbs to $1.5 trillion. Against that, the entire $304 billion of lifetime private capital raised by the NatSec100 cohort — impressive as it sounds — is a rounding error, and the annual flow of federal awards to these companies is smaller still. Private capital is financing the industrial base that the customer has not yet agreed to buy at scale. That is a bet, not a business model, until the production share keeps climbing.
The concentration is another fragility. The $14.6 billion raised in the first five months of 2026 is not a broad market surge. Anduril's $5 billion round alone accounts for more than a third of it. A tracker that counted 235 defense-tech financings across 172 companies from January 2024 through September 2026 found a median round of just $15 million against an average of $123.3 million — a distribution skewed hard to the right by a handful of mega-deals. Only 26.4 percent of the rounds were larger than $50 million, yet those deals captured 91.9 percent of all the capital recorded. Half of all cumulative defense-tech funding sits with the top five companies; around 80 percent sits with the top twenty.
This is where the stardust metaphor earns its weight. Stardust is what glitters after an explosion — fine, dispersed, and already spent. Much of what is being celebrated is the glow of a financing event, not the durability of a funded program of record. Ninety-two percent of awards to these companies are still not production. A contract vehicle is not revenue; an IDIQ ceiling is not a task order.
"Product quality and reliability no longer define success for non-traditional companies," said James Cross, co-founder and board member of the Silicon Valley Defense Group. "Instead, these companies must also ensure they can deliver production at volume."
The Strongest Counter-Case: A Shakeout, Not a Supercycle
The most credible bear case does not come from a skeptic of defense spending. It comes from inside the movement. Trae Stephens, Anduril's executive chairman, warned in June 2026 that "defense tech is headed for a shakeout." The logic is straightforward: the capital that arrived in 2025 and 2026 was priced for a world in which procurement reform continues, conflicts remain hot, and the Pentagon converts its new contract vehicles into funded, multi-year task orders. If any one of those assumptions fails, the secondary-market marks — Anduril at $95 billion, Shield AI at $10 billion on implied secondary trading — have no revenue anchor capable of supporting them.
The market has already offered a preview. Of the eleven defense companies that went public in 2025 and 2026, seven trade below their IPO price, most down roughly 50 percent. Only a handful — Swarmer, Arxis and Karman — are up, by 60 to 96 percent. The public market is not yet rewarding defense-tech IPOs; it is punishing most of them.
The historical precedent is not comforting either. Defense-tech funding has always been lumpy; a single large round can swing the industry-wide number by tens of percentage points. Through the middle of the second quarter of 2024, for instance, defense startups had raised just $228 million, a 62 percent decline from the nearly $600 million raised at the same point in 2023 — a drop attributed to the timing of a few giant rounds the prior year. The median company in this cohort is not raising billions. It is raising $15 million a round, and most of that money is still financing prototypes.
The counter-thesis, stated at its strongest: this is a valuation bubble riding a war-urgency wave, not an industrial revolution. The evidence is the 92 percent non-production award share, the seven-to-one private-to-government funding ratio, and the widening gap between private capital and government revenue that the ecosystem's own report flags as a systemic risk. If the wars in Ukraine or the Middle East freeze, or if Congress funds the defense topline but not the new vendors, the companies without factories and funded programs of record will not survive to become the new primes.
That counter-case is serious, but it mistakes the nature of the shift. The taboo is gone structurally, not cyclically: Google rewrote its AI principles rather than reverting to the 2018 position; dedicated defense funds now exist as a permanent allocation class; procurement pathways that took decades to build are now law rather than aspiration. A shakeout will happen — it always does after a capital surge — but a shakeout is how a category matures, not how it dies. The survivors of a shakeout become the index.
The falsifying signal is quantifiable. If the share of NatSec100 awards that are production contracts does not rise above roughly 20 percent within two years, and if aggregate federal awards to these companies remain below 2 percent of the defense budget, then the structural-adoption thesis is wrong and this was a war-urgency valuation bubble. Watch those two numbers, not the funding headlines.
What Comes Next: Three Time Horizons
Short term (6–18 months): momentum continues. The FY2026 and FY2027 defense budgets swell the topline, war urgency keeps procurement timelines compressed, and more "largest ever" contract headlines arrive. Secondary valuations will keep climbing as long as primary rounds clear at higher marks. Palantir's guidance — U.S. commercial revenue expected to grow at least 143 percent in 2026 — signals that the commercial spillover from defense-proven AI is only beginning.
Medium term (2–4 years): the shakeout. The mega-rounds of 2025–2026 cannot be repeated at the same cadence. Companies that cannot convert IDIQ vehicles into funded task orders, or that cannot build the manufacturing capacity to fulfill production contracts, will run out of runway. Mergers, acquisitions by the legacy primes, and the first wave of public listings will separate the category into survivors and casualties. This is the phase where the stardust settles and you can see who is still standing.
Long term (5+ years): bifurcation. If procurement reform holds and production share keeps rising, the survivors become a new tier of prime contractor alongside Lockheed, Northrop and Raytheon, and defense becomes a permanent, non-taboo allocation in venture portfolios. If reform stalls, the sector reverts to a niche, and the $61 billion valuations become a cautionary footnote about how fast capital can mistake urgency for demand.
Three scenarios frame the path. In the base case, Anduril, Palantir, Shield AI and Saronic scale into production, the production-contract share reaches roughly 20 percent by 2028, and two or three NatSec100 companies go public or are acquired by primes. In the upside case, conflicts persist or expand, the $1.5 trillion FY2027 request passes, reform accelerates, and several $50 billion-plus defense-tech companies emerge. In the downside case, the wars freeze, Congress funds the topline but not the new vendors, IDIQs stay unfunded, and secondary valuations halve in a shakeout that wipes out half the cohort.
The signal to watch is not the next funding round. It is the next production contract. Funding tells you what investors believe; production tells you what the customer has decided to buy at scale. Everything else is atmosphere.
The stardust is real — but stardust is what settles on a battlefield after the explosion, and most of these companies have not yet built anything that survives contact with the Pentagon's procurement machine. The ones that do will not need the glow. The ones that don't will be remembered as the moment defense tech looked, briefly, like the next software.
Explore more exclusive insights at nextfin.ai.
