$14.6B and Counting: Why Defense Tech Is Drawing the Biggest Checks in Venture

$14.6 billion. That's how much defense tech startups raised in just the first five months of 2026 — already surpassing the $9.6 billion pulled in across all of 2025. The gap is widening at a pace that should make everyone in the defense industrial base stop and look. Anduril closed a $5 billion Series H at a $61 billion valuation. Shield AI, Saronic, and a dozen others closed nine-figure rounds. And — critically — it wasn't just specialist defense funds writing those checks. Thrive Capital, Andreessen Horowitz, and Sequoia were on the cap tables. Mainstream money, writing real checks, into a market that most of Silicon Valley spent a decade ignoring. Something structural has changed.

It's Not Just the AI Tailwind

The easy explanation is that AI fever has infected defense spending — that investors are simply riding the same autonomous-systems narrative that has driven AI investment across every other sector. That's partially true, but it misses the more important shift underneath. The investors who are piling into defense tech aren't buying the AI narrative in isolation. They're buying something more specific: the idea that battlefield software and sensor networks are compounding platforms, not one-off weapons contracts.

That distinction matters enormously. A one-off weapons contract has a finite ceiling — you sell the system, the procurement ends, and the relationship resets for the next program. A compounding platform gets more valuable as it accumulates data, learns from operational use, integrates with more systems, and becomes harder to displace. Investors see sensor networks, autonomous-software stacks, and edge-compute infrastructure in the defense market as possessing exactly those compounding properties. The unit economics — long a concern for defense tech — finally work.

The implications are significant for how the market will evolve. Companies that can demonstrate compounding network effects — where each deployment makes the next deployment more capable — will command valuation premiums that pure hardware companies cannot match. The defense market of 2030 will look less like Raytheon and more like Palantir: software-defined, data-accumulative, and priced accordingly.

The Program Office Is Eating the Prime

One of the most underappreciated dynamics in the current funding surge is the degree to which the traditional prime-contractor model is being circumvented. For decades, the path to defense revenue ran through a handful of companies that controlled access to program offices, held the relevant security clearances, and could navigate the acquisition bureaucracy. That moat is not gone, but it has been significantly narrowed.

The CCA program's deliberate multi-vendor software architecture, OTA contracts, and Other Transaction Authorities are examples of a broader shift: the Pentagon is actively seeking direct relationships with technology companies that can deliver capability without requiring them to route through a prime. The Defense Innovation Unit, the Air Force's AFVentures, and the Army's xTech program have collectively created pathways that didn't exist five years ago.

For startups and growth-stage companies, this means that the barrier to defense revenue is lower than it has ever been — not low, but meaningfully lower. The constraint is no longer a prime's willingness to partner. It's whether a company can demonstrate operational performance, pass security vetting, and build a production pipeline. Investors are pricing in the possibility that the next decade will see a much larger share of defense spending flow directly to technology companies, bypassing the traditional industrial base entirely.

Where the Smart Money Is Going

The funding data makes the investment theses fairly clear. Autonomous systems — drones, unmanned platforms, autonomous wingmen — has absorbed a significant share of capital. So has AI and software for defense applications: battlefield awareness, decision-support tools, autonomous target recognition. And sensor networks — the layer that feeds autonomous platforms and human decision-makers alike — are attracting substantial investment.

What's notable is the concentration in areas that serve the edge: deployed, forward-operating, often contested environments where centralized infrastructure cannot be assumed. This reflects a strategic consensus — inside both the Pentagon and the investment community — that the next conflict will be decided at the tactical edge, not in rear-area command centers. Platforms and systems that operate effectively without relying on stable communications, continuous GPS, or centralized processing are the ones investors believe will see sustained procurement.

The corollary is that defense technology investment is increasingly being valued on its operational characteristics, not just its technical specifications. A sensor that works in a laboratory is interesting. A sensor that works in an urban environment, through walls, without emitting detectable signals, at a price point compatible with attritable platforms — that's a platform-compounding asset. Investors are getting sophisticated about the distinction.

Is the Prime Model Broken?

The defense primes — Lockheed Martin, Raytheon, Northrop Grumman, Boeing — are not unaware of what's happening. Anduril's $61 billion valuation compares favorably to the enterprise values of some legacy primes. Shield AI is competing directly with established players for autonomous-systems contracts. The question of whether the new entrants will simply get absorbed by the primes, or whether they represent a durable structural shift, is one of the most important strategic questions in defense technology right now.

The optimistic case for the primes is that they have deep security credentials, proven manufacturing scale, and relationships that took decades to build. The pessimistic case is that those same attributes make them slow, expensive, and culturally unable to attract the engineering talent that builds compounding software platforms. The truth is somewhere in between — and the most likely outcome is a restructuring of the defense industrial base in which primes acquire or partner with technology companies to stay relevant, while the most capable independent companies pursue direct-to-Pentagon pathways that the primes cannot easily replicate.

For investors, the bet is that the independents win more often than they lose. The funding numbers suggest that bet is being placed with real conviction.

What This Means for Sensing Companies

The capital environment creates a specific set of opportunities for companies building sensing technology for defense applications. The funding surge signals that investors are willing to back sensing platforms as strategic assets — not merely as components within a larger system — provided those platforms demonstrate compounding network properties and edge-readiness.

Three factors determine whether a sensing company captures that opportunity or gets passed over:

• Operational deployment, not just demonstration. The defense market has seen enough impressive prototypes. Investors and procurement officers alike are looking for systems that have been tested in operational conditions — not just in controlled environments. Sensing companies that can point to real-world deployment data have a significant advantage in a market where credibility is the primary bottleneck.
• Edge-native architecture. Sensing systems that require centralized processing, stable comms, or precise GPS are poorly matched to the environments where autonomous platforms will operate. The sensing layer that attracts investment is one that processes at the edge, communicates minimally, and degrades gracefully under contested conditions. Passive RF sensing — detecting human presence and movement through walls using ambient WiFi signals, without emitting anything detectable — is architecturally suited to exactly this requirement.
• Integration pathways, not just technology. A sensing capability that exists in isolation is a science project. The companies that attract meaningful investment are those that have mapped their technology to specific program needs, have identified the right entry points in the acquisition pipeline, and can articulate how their sensing layer multiplies the value of the autonomous platforms and decision-support systems it feeds.

Ceradon's Take

The $14.6 billion figure is a number, but what it represents is more important: a structural re-rating of the defense technology market by investors who spent the 2010s on the sidelines. The defense tech category has been demystified. The technical risk is understood. The procurement pathways are navigable. And the market size — defined broadly as the sensing, software, and autonomous systems needed for a distributed, edge-centric battlespace — is an order of magnitude larger than what the traditional defense industrial base was designed to address.

For companies building in this space, the funding environment is a tailwind but not a moat. Capital is abundant; conviction is cheap. What separates the companies that use this moment to build durable businesses from those that ride the wave and capsize is the same as it's always been: whether they can deliver operational performance at the edge, at scale, on the timelines that the current strategic environment demands.

The sensing layer is where Ceradon operates — passive WiFi CSI through-wall detection that serves the edge-native, autonomous-ready, attritable-platform paradigm that investors and program offices alike are now prioritizing. The capital is flowing. The programs are standing up. The question for every company in this space is whether they can execute at the pace the moment requires. That's the only bet that actually matters.