Notes from the Frontier
What defense tech founders know that your banker doesn't
In August 2017, a company called Anduril Industries raised $17.5 million in seed funding at a valuation of roughly $88 million. The founding team included a fired Facebook executive, a Palantir alumnus, and a handful of engineers with defense backgrounds but no defense contracts. There was no revenue. There was no product in market. What there was, in the words of executive chairman Trae Stephens, was a thesis: that the United States Department of Defense would eventually buy software-defined autonomous systems from startups, and that whoever built the operating system for those systems first would own the category.
Eight years later, Anduril is valued at $30.5 billion. It doubled its revenue to $1 billion in 2024. Founders Fund, which led the seed round, wrote its largest check in firm history to lead the company's $2.5 billion Series G in June 2025.
The question for family office allocators is not whether Anduril is an extraordinary company. It is when, exactly, they first heard about it. For most, the answer is sometime around the Series D or Series E, when the valuation was already between $4.6 billion and $8.5 billion and the company had appeared in enough bank research notes to seem investable. By then, the seed-stage thesis had been validated many times over. The informational edge had long since been priced in. The entry point was comfortable. It was also late.
This is the asymmetry that matters most in private markets today. It is not the gap between one asset class and another, or between one vintage and the next. It is the gap between what frontier founders know and what their eventual investors learn, and the months or years of compounding value that disappear into that gap.
A $2.7 trillion demand signal
To understand why this asymmetry is accelerating, start with the demand side. Global military expenditure reached $2.7 trillion in 2024, according to the Stockholm International Peace Research Institute. The 9.4 percent year-over-year increase was the steepest since the end of the Cold War. Spending rose in every region. Eighteen of thirty-two NATO members met or exceeded the alliance's 2 percent of GDP guideline, the highest number since the target was adopted in 2014. Germany's defense spending surged 28 percent in a single year, making it the largest military spender in Western Europe for the first time since reunification.
The capital flowing into defense technology startups reflects this structural shift. Venture capital investment in defense tech hit $49.1 billion in 2025, according to PitchBook, nearly doubling from $27.2 billion in 2024. Equity funding alone more than doubled to $17.9 billion, per CB Insights. The number of firms actively investing in defense tech rose 41 percent in a single year, as mainstream venture firms dropped previous ethical objections and reframed defense investing as supporting democratic values.
These are not speculative flows chasing a theme. They are capital responding to procurement budgets that have already been authorized. The United States has committed to a defense budget approaching $1 trillion. The European Union is planning up to 800 billion euros in defense spending by 2030. When governments move this much money this fast, the question for private capital is not whether to participate but how to gain access to the companies that will absorb it.
When the Porsche family starts a defense fund
The shift is visible at every level of the private capital ecosystem. In August 2025, Porsche Automobil Holding SE, the holding company controlled by the Porsche-Piech family, announced it would build a platform for investments in defense technology startups. In November 2025, Porsche SE hosted a "Defense Day" event for German and European family offices interested in the sector. According to Bloomberg, the initiative targets approximately half a billion euros in venture capital, with Deutsche Telekom also set to participate. Porsche SE already holds positions in dual-use technology companies including Isar Aerospace and Quantum Systems.
The Porsche-Piech move is instructive because of what it reveals about the information problem. Here is a family with multigenerational wealth, deep industrial expertise, and existing relationships across European manufacturing, and it still needed to create a dedicated event and investment platform to bridge the gap between its capital and the defense technology ecosystem. If one of Europe's most connected industrial families requires a purpose-built structure to access the sector, the median single-family office with four investment professionals has no chance of sourcing these deals through conventional channels.
The pattern is the same in the United States. When former Google CEO Eric Schmidt, Stanley Druckenmiller, and Laurene Powell Jobs all participated in an $863 million round for nuclear fusion company Commonwealth Fusion Systems in 2025, they were not responding to a pitch deck circulated by a placement agent. They were acting on relationships built over years of direct engagement with founders building at the intersection of energy infrastructure and national security. Barry Sternlicht's family office, Jaws Ventures, joined a $50 million round for Zeno Power, a startup producing compact nuclear batteries for the Department of Defense and NASA. Jeff Bezos' family office backed Atlas Data Storage and robotics startup FieldAI. These are investments that never appeared on a standard deal-flow calendar because they were never distributed through standard channels.
The knowledge that doesn't travel
The information asymmetry in deep tech is qualitatively different from the asymmetry in conventional venture. In consumer technology or SaaS, an investor can evaluate a startup by examining market size, user growth, unit economics, and competitive positioning. The data is accessible. The frameworks are portable. A generalist allocator can get reasonably close to the right answer with a spreadsheet and a few reference calls.
In defense technology, nuclear energy, and AI infrastructure, the relevant knowledge is technical, classified or export-controlled, and embedded in relationships that do not transfer through traditional intermediaries. A founder building autonomous naval vessels knows which NATO procurement programs are opening, which prime contractors are losing confidence in their existing suppliers, and which specific capability gaps remain unfilled after decades of legacy program delays. A founder developing microreactors for military deployment knows the regulatory pathway through the Department of Energy's Reactor Pilot Program, the timeline for HALEU fuel availability, and the specific thermal requirements of forward-deployed data centers. None of this appears in a Morgan Stanley research note.
The result is that the most consequential investment opportunities of the current cycle are being originated by people who understand the technical landscape, sourced through direct founder relationships, and closed before the broader market develops a consensus view. Point72 Ventures, an active defense tech investor, observed that much of the sector's legacy industrial base was designed for an era of six-to-eight-year timelines and hardware-first thinking. The startups displacing that base operate on fundamentally different assumptions. Understanding those assumptions requires proximity to the founders making them.
The infrastructure gap
This is precisely where the operational data from earlier in this series intersects with the investment thesis. Goldman Sachs found that the typical family office investment team consists of fewer than five people. BlackRock's 2025 survey revealed that 63 percent of family offices acknowledge gaps in deal sourcing, 75 percent in private-market analytics, and 57 percent in reporting. These are not abstract deficiencies. They are structural barriers to accessing the categories producing the highest-conviction returns.
A four-person team cannot simultaneously manage an existing portfolio, evaluate conventional fund commitments, and develop the technical fluency required to originate direct investments in autonomous systems, advanced nuclear, or AI infrastructure. The bandwidth constraint is absolute. And the consequence is predictable: family offices that lack dedicated sourcing infrastructure default to the deals that come to them through banks, conferences, and GP relationships. Those deals are, almost by definition, the ones that have already been widely distributed.
The BNY Wealth survey found that 64 percent of family offices plan to make six or more direct investments in the coming year, and that alignment of interests saw a 52 percent year-over-year increase as a priority consideration. The intention is there. What is missing, for most, is the infrastructure to act on it in the sectors where it matters most.
Proximity as an investment function
The most significant implication of this asymmetry is organizational, not financial. The offices that captured early positions in defense tech and hard tech did so not because they had better financial models but because they had better networks. They treated founder relationships as an intelligence-gathering function. They attended the right working groups, hired analysts with technical backgrounds, and positioned themselves as capital sources that understood what the founders were building before the pitch deck existed.
A founder who recently chose a European family office as lead investor over a tier-one venture firm put it simply in an interview with CNBC: "Stop caring about the name of your investor. If things go wrong, you want someone that's going to spend time with you as opposed to just saying, I'm going to write off this company and focus on the next big one."
That preference is a structural advantage for family offices willing to do the work. Patient capital, long time horizons, no quarterly redemption pressure, and the ability to hold assets over multiple generations: these are precisely the attributes that deep-tech founders need in their cap tables. Goldman Sachs' Meena Flynn noted that family offices have the ability to invest in assets they can hold over multiple generations without worrying about an exit. The alignment is natural. What is unnatural is the distance that currently separates the capital from the opportunity.
The Q2 thesis continues
This article is the second in a Q2 series on conviction capital. The first examined the structural shift from spray-and-pray allocation to concentrated, thesis-driven investing. This piece argues that conviction is necessary but insufficient. Without the sourcing infrastructure, technical fluency, and founder relationships to access non-consensus opportunities before they become consensus, conviction capital will continue to arrive late and pay full price.
The offices getting this right treat deal origination as a first-order operational function, not a byproduct of conference attendance. They are investing in the capabilities that close the 63 percent deal-sourcing gap identified by BlackRock. And they are doing so at a moment when the sectors generating the most asymmetric returns require the most specialized access.
The frontier is not a metaphor. It is a specific set of technologies, a specific set of founders, and a specific set of relationships. The allocators who are proximate to that frontier will capture the next decade of compounding value. The rest will read about it in a research note, two or three funding rounds too late.
This is the seventh installment of The Prominent Blog, a biweekly series on the convergence of capital strategy and operational technology in the family office sector.
Sources and verification notes:
All statistics cited in this article are drawn from identified, published sources:
Stockholm International Peace Research Institute (SIPRI), Trends in World Military Expenditure, 2024 (published April 28, 2025): Global military expenditure reached $2.718 trillion in 2024, a 9.4% year-over-year increase, the steepest since the end of the Cold War. Spending increased in all five geographic regions. 18 of 32 NATO members met or exceeded the 2% of GDP guideline. Germany's military spending rose 28%, making it the largest spender in Central and Western Europe for the first time since reunification. Confirmed via SIPRI Fact Sheet, SIPRI press release, CNN (Apr 30, 2025), France24 (Apr 28, 2025).
PitchBook 2025 Vertical Snapshot: Defense Tech (August 2025): VC defense tech deal value reached $49.1 billion in 2025, nearly doubling from $27.2 billion in 2024. Median VC defense tech valuation in 2025 was $146 million, up from $42.8 million in 2024. Number of firms actively investing in defense tech rose 41%. Confirmed via PitchBook reports, Defense News (Jan 20, 2026), Tectonic Defense (Dec 23, 2025).
CB Insights: Equity funding for defense tech startups more than doubled to $17.9 billion in 2025 from $7.3 billion in 2024. Defense tech outpaced overall equity funding growth (47%). Confirmed via Defense News (Jan 20, 2026).
Crunchbase: VC-backed defense startups (military, national security, law enforcement) raised $7.7 billion across ~100 deals in 2025, more than double 2024's $3.2 billion. In 2024, $3 billion across 102 deals, an 11% uptick from 2023. Confirmed via Crunchbase News (Nov 26, 2025; Mar 26, 2025).
Anduril Industries funding history: Seed round $17.5 million (August 2017); Series A $41 million at ~$250 million valuation (June 2018); Series B at ~$1 billion valuation (2019); Series G $2.5 billion led by Founders Fund at $30.5 billion valuation (June 2025); revenue doubled to $1 billion in 2024. Confirmed via TechCrunch (Jun 5, 2025), CNBC (Jun 5, 2025), Fortune (Jun 6, 2025), Forge Global, Contrary Research, Sacra.
Porsche Automobil Holding SE: Announced defense investment platform and "Defense Day" (August 2025); event held November 5, 2025 for German and European family offices. Approximately half a billion euros in venture capital targeted; Deutsche Telekom also set to participate. Existing dual-use investments include Isar Aerospace and Quantum Systems. Confirmed via Porsche SE press releases (Aug 13, 2025; Nov 11, 2025), Heise Online, Pensions & Investments (Aug 13, 2025), Jalopnik (Aug 15, 2025).
Family office deep tech investments: Eric Schmidt, Stanley Druckenmiller, Laurene Powell Jobs, and Bill Gates (Gates Frontier) participated in $863 million Series B2 for Commonwealth Fusion Systems (August 2025); Barry Sternlicht's Jaws Ventures joined $50 million round for Zeno Power (May 2025); Jeff Bezos' family office backed Atlas Data Storage ($155 million seed) and FieldAI ($314 million). Confirmed via CNBC Inside Wealth (Jun 5, 2025; Sep 4, 2025).
Goldman Sachs 2025 Family Office Investment Insights ("Adapting to the Terrain"): 245 institutional family offices; investment teams typically fewer than five people; 86% invest in AI; 58% plan to overweight technology. Meena Flynn quote on holding assets "over multiple generations." Confirmed via Goldman Sachs press release (Sep 10, 2025), Goldman Sachs article (Oct 15, 2025), CNBC (Sep 11, 2025).
BlackRock 2025 Global Family Office Survey: 175 single-family offices; 63% acknowledge gaps in deal sourcing; 75% in private-market analytics; 57% in reporting. Confirmed via BlackRock press release (Jun 2025), BusinessWire, Nasdaq.
BNY Wealth 2025 Investment Insights: 282 SFOs; 64% plan 6+ direct investments (10% increase); 52% YoY increase citing alignment of interests as crucial. Confirmed via BNY report, Family Wealth Report (Nov 17, 2025).
Point72 Ventures on defense sector legacy industrial base and six-to-eight-year timelines. Confirmed via Crunchbase News (Nov 26, 2025).
Founder quote ("stop caring about the name of your investor"): Benkirane, founder of Spore.Bio, on choosing Smedvig Ventures (family office) over tier-one VC. Confirmed via CNBC Inside Wealth (Mar 7, 2025).
EU defense spending target: Up to 800 billion euros by 2030. U.S. defense budget approaching $1 trillion. Confirmed via PitchBook, Techloy (Dec 4, 2025), SIPRI.
