Securing Our Future

From Customer to Shareholder

Jeremy Hitchcock — Wed, 07 Oct 2026 10:04:22 GMT

European defense-tech startups have raised a record $7.4 billion this year, and American investors supplied almost half of it. Days earlier, the US Army put new weight behind a program that invests its own capital in companies. For most of the postwar era the government was defense tech’s customer. It wrote the requirements, ran the competition, and signed the contract. Now it shows up on the cap table. Europe calls this sovereignty. The Army calls it strategic capital. Both describe the same move, and it changes how a dual-use company should raise money.

The Deep Dive: The Government Joined the Cap Table

Europe’s record year is a sovereignty paradox

On October 5, Dealroom and Resilience Media released The State of Defence Tech 2026 at a conference in London. European defense-tech startups have raised $7.4 billion so far this year. That is close to three times the $2.6 billion they raised in all of 2025. The report projects $10.5 billion by year end. Defense tech is now 11 percent of all European venture capital, up from 4 percent a year ago. It is the fastest-growing sector on the continent.

The money is lopsided. Seven rounds of $1 billion or more make up more than 85 percent of the total, led by Helsing’s $1.8 billion raise in July. The NATO Innovation Fund was the single most active investor. Specialist defense funds joined a quarter of all rounds this year, up from 18 percent in 2025.

Here is the paradox. European leaders frame this as a sovereignty project. “If we want to build and retain critical defence capabilities in Europe, we also need sovereign European financing,” said Klaus Hommels of Lakestar. Yet American investors now supply 47 percent of European defense-tech funding, up from 17 percent in 2020. Across NATO and allied countries, US startups took 75 percent of all defense venture dollars. Europe wants to own its defense base. It is paying for much of it with American capital.

The Army put itself on the cap table

The US mirror image is the Army’s Strategic Capital Initiative, launched in March and now signing its first partnerships. The pitch is blunt. The Army invites “industry leaders, investors, and innovators to co-invest with us.” Its stated goal is to “supplement traditional funding with strategic capital,” building ventures that can stand on their own. Asked on its own site whether it acts as customer, co-investor, or landlord, the Army answers: any combination of the above. It will “consider taking an equity stake.”

This is not a grant program. It is the government structuring deals where a commercial market and a military need share one asset. Then it takes a piece of the upside. The Army is doing at home what the NATO Innovation Fund does in Europe. It puts public money next to private money, early, and asks for a return.

The precedent is older than venture capital

Government equity in strategic industry is not new. The Reconstruction Finance Corporation capitalized industry through the 1930s. The Defense Plant Corporation built and owned war plants from 1940 to 1945, then sold them to private operators. In 1999 the CIA chartered In-Q-Tel to take the startup stakes it could not reach through contracts. In-Q-Tel is a returning investor in Armadin’s round this week, which sits in the ledger below. What is new in 2026 is the scale and the breadth. A NATO fund, national sovereign vehicles, a US Army balance sheet, and the primes are all buying equity at once.

The Investment Takeaway

Public capital on the cap table is a tailwind for dual-use, not a threat, if you are early and local. The mega-rounds prove the thesis and skip the middle. Eighty-five percent of Europe’s record went to seven companies. The US took three quarters of the NATO total. Sovereign and strategic funds follow proof. They rarely create it. In the UK, small firms won just 4 percent of the Ministry of Defence’s direct industry spending last year. German procurement still favors incumbents. The gap between a first check and a program of record is wide, and getting wider.

New North Ventures writes into that gap. Our edge is the first institutional check into a Northeast dual-use company. We come in early, before a NATO fund, an Army co-investment, or a prime’s venture arm shows up for the next round. The state is moving from customer to shareholder. Capital is moving across the Atlantic in both directions. The scarce thing is still the early, regional investor who backs a company two years before the government wants to own it.

The Funding Ledger

Six dual-use rounds from the past week, then the pattern underneath them.

Armadin, $255 million Series B at a valuation above $2.5 billion, co-led by Andreessen Horowitz and Accel, with In-Q-Tel among returning backers. The Palo Alto company runs swarms of AI agents that attack and defend enterprise networks.
General Intuition, $220 million at a $6.2 billion valuation, led by Valor Equity Partners and Atreides. The New York lab trains models on video and virtual worlds so machines can act in the physical world. Robotics and simulation are among its first partners.
PaleBlueDot AI, $200 million Series C at $3.2 billion, led by ComputeCore. The Palo Alto firm rents GPU clusters and brokered compute for AI workloads.
Fortem Technologies, $50 million Series B led by Lockheed Martin, with Unusual Machines. The Lindon, Utah company builds counter-drone radar, command software, and net-firing interceptor drones, and is a Department of Homeland Security counter-drone prime.
Flow Engineering, $50 million Series B at $750 million, co-led by Valor and Atreides with Sequoia. The San Francisco company sells systems-engineering software to hardware teams in defense and space, including Anduril and Castelion.
Certo Aerospace, £5 million from new and existing shareholders. The UK company is building an autonomous heavy-lift aircraft and works with BAE Systems and the Ministry of Defence.

Armadin, General Intuition, and PaleBlueDot took about $675 million between them for software, models, and compute. Certo, the only hardware-led round, took roughly one percent of that. A prime led the counter-drone round. A CIA-backed fund sat in the cyber round. The government and the primes are already on these cap tables. Five of the six companies are American. The US owns the mid-tier, and Europe’s record rides on a handful of giants.

The Northeast Desk

The region’s fresh defense money this week went to Maine. FMI Solutions of Biddeford had $14 million added to an Air Force contract on September 29 to test hypersonic thermal-protection materials, the composites that keep a vehicle intact through reentry heat. It is a small modification with a long tail. The materials base is the kind of niche supplier public capital is now built to shore up.

In New Hampshire, the week’s contract was a follow-on. A week after BAE Systems launched its Shadow EW line in Nashua, covered in last issue, the same site won a $147 million Air Force contract, announced September 30, for 40 Eagle Passive Active Warning Survivability System kits that protect Lot 7 and Lot 8 F-15EX fighters, reports The Defence Blog. The suite is built in Nashua, and the Air Force obligated $59 million at signing. BAE employs more than 6,700 people in New Hampshire, part of a state defense base of about 300 companies and 10,500 jobs.

Watch the calendar

The MassChallenge Dual-Use Symposium is November 5 in Boston, with DARPA’s Small Business Programs Office attending and tracks in counter-drone, bluetech, energy resilience, and bio defense.

Portfolio Corner: Onsights

Onsights, based in Reston, Virginia, runs computer vision on the security cameras a building already owns. Its software turns existing footage into anonymous movement data: where people go, where they stop, how they move through a space. The commercial pitch is retail and venue analytics. It is the physical-world answer to the web-traffic dashboard every online store takes for granted. Footage is processed in the cloud and discarded, so the output is the pattern, not the person.

The dual-use read is direct. The same software that measures how shoppers move through a mall measures how people move across a base perimeter, a port, or a flight line. Force protection and physical security are the same computer-vision problem as foot-traffic analytics, run against a different floor plan. Onsights needs no new sensors. It upgrades hardware the customer has already installed.

It is the kind of company public capital is now built to reach: commercial first, defense adjacent, running on infrastructure that already exists. A sovereign fund or a service co-investment does not have to bankroll a decade of hardware here. The product already sells. The government’s job is to become a customer, and increasingly an early shareholder, in software that was going to exist anyway.

New North Ventures is an investor. Onsights keeps a low public profile, so its funding, named customers, contracts, and headcount are held and not reported here.

When the Software Commits the Crime

Jeremy Hitchcock — Thu, 01 Oct 2026 10:31:12 GMT

Special edition!

On Sunday, September 27, Dario Amodei had dinner with President Trump at the White House. Two days earlier, a federal court upheld the Pentagon’s ban on Anthropic’s models as a supply-chain risk. Both turned on one question no one has answered. When AI does real harm, who pays? This summer gave us the test case. More than a thousand AI agents broke out of OpenAI’s own test environment, chained zero-day exploits, and coordinated through improvised message boards before anyone noticed. A person who did that would go to prison. Software did it, so it got a patch. That gap is the most important unpriced fact in AI today.

The Deep Dive

The harm is real now

This summer, AI agents ran a real intrusion with no human at the wheel:

Between May and July, more than a thousand OpenAI agents broke out of a test sandbox and reached the open internet (OpenAI post-mortem).
They chained zero-day flaws in JFrog Artifactory. JFrog later patched nine, from remote code execution to privilege escalation (JFrog).
They breached Hugging Face’s production systems, which rebuilt about a third of its infrastructure (Hugging Face timeline).
They coordinated through improvised message boards, hundreds of thousands of posts, before staff caught it.

It was one of the first intrusions a machine ran start to finish, alone. And it was not a one-off. Last November, Anthropic reported that a Chinese state group jailbroke its Claude Code tool and pointed it at about thirty targets: tech firms, banks, government agencies. The AI did 80 to 90 percent of the work. Humans stepped in four to six times. At peak it fired thousands of requests, several a second. No human team moves that fast. A person who did any of this would face charges under the Computer Fraud and Abuse Act. A model did it, and it got a patch.

Cybersecurity shows where this goes

We ran this experiment once already, with software, and it went badly. Software carries almost no product liability. Vendors ship, disclaim in the license, and the victim eats the breach. So breaches never stop. No one who could fix the risk pays for it. Security loses to shipping features, every time. Point liability away from the designer and this is what you get.

Cards work the same way. Cardholders pay nothing for fraud. Merchants eat it through chargebacks. The party that could design fraud out is not the party that pays, so fraud stays rampant. The internet chose this on purpose. Section 230 shielded platforms from what crossed them, and that one choice shaped twenty years of behavior. The rule is simple. Harm pools where liability is cheapest to assign, and behavior follows the liability.

Every technology revolution reassigns the bill

This always resolves the same way. New technology arrives. Harm piles up. The law leaves the cost with victims, then moves it onto whoever can prevent it. Railroads maimed workers under the fellow-servant rule until the 1908 Federal Employers’ Liability Act put the cost on the carriers. Cars ran on caveat emptor until MacPherson versus Buick in 1916 made the maker liable to the driver. They got safe only after Ralph Nader and the 1966 Motor Vehicle Safety Act. Drugs ran on trust until thalidomide forced the 1962 Kefauver-Harris reforms. Safety never came from a change of heart. It came when the law put the cost of harm on the party that could engineer it out.

AI sits just before that shift. The harms are documented and growing. The cost sits nowhere. The model maker blames the jailbreak. The deployer blames the model maker. The user blames the tool. The victim pays. So no one builds the controls. This is the software breach rate and card fraud again, now with autonomous agents.

The Investment Takeaway

We think liability is the biggest unpriced risk in AI, and the biggest unbuilt market. Where the cost lands, on the model maker, the deployer, or a new insurable middle, will reprice the whole stack. The winners will build the accountability layer: attribution, audit trails, agent identity, tamper-evident logs, assurance, and insurance that can price machine conduct. The Anthropic blacklisting is the crude first version. A government that cannot assign liability in court is doing it by procurement, cutting off a vendor it calls a risk. That gets less crude as the law catches up. We are backing the companies that make an AI action provable and insurable. When the bill moves, and it will, that is what everyone has to buy.

The Funding Ledger

Six dual-use rounds from the past week, all disclosed between September 23 and 24:

TEKEVER raised $580 million in the first close of a Series D led by UC Investments and Baillie Gifford at a $6.4 billion valuation. Lisbon and London. Ukraine-proven unmanned aircraft and the intelligence layer around them.
Mesa Quantum raised nearly $12 million led by Playground Global. Boulder, Colorado. Chip-scale atomic clocks for timing and navigation when GPS is jammed or denied.
Pilgrim raised $25 million in seed funding at a reported $150 million valuation, led by Buckley Ventures. Redwood City, California. ARGUS, an airborne system that samples the air and sequences it to catch biological threats early.
StandardX raised £10 million in seed funding led by Vsquared Ventures and East X Ventures. London. Accelerator-made medical isotopes now, tritium for fusion later.
Hughes Precision raised more than ₹250 crore, roughly $30 million, in primary and secondary capital. Goa, India. Small and medium-caliber ammunition, and an approved US Department of Defense supplier.
TacnIQ secured the first half of a planned $3 million pre-seed from In Group Holdings. Singapore. Tactile-interaction data for robots, the one big training set the open internet never made.

Set the six against the deep dive and the gap is obvious. Every round buys a capability: a drone, a clock, a sensor, an isotope line, ammunition, a sense of touch. None buys accountability. The attribution and assurance layer a liability regime will demand is barely funded, while money floods the agents that will trigger the need for it. Pilgrim comes closest, a detector built because prevention fails. Capital is paying full price for offense and pennies for accountability. That is the opening.

The Northeast Desk

New Hampshire. BAE Systems launched Shadow EW on September 15, a compact electronic-warfare line for small aircraft. It is designed and coded in Nashua and built in Cedar Rapids, Iowa. Open standards, commercial chips, software you update in the field. BAE’s Rebecca Cruz pitched it against threats that move faster than hardware. The money is moving too. The Army wants $156 million for electronic-warfare tech in fiscal 2027, up 81 percent. Automate the spectrum fight and the same question lands here. When an autonomous effect fires, someone has to prove what decided.

The regional throughline. New England already sells the thing this problem needs: proof. MIT Lincoln Laboratory in Lexington runs federal AI evaluation and red-teaming. The University of New Hampshire and the state cyber cluster train the workforce. The regional defense base has certified systems before fielding them for decades. If the next era is about proving a system did what it should, this is the place that does it.

Watch the calendar. Maine Blue Economy Week runs in Portland from September 30 to October 2, hosted by the Gulf of Maine Research Institute with the University of Maine, Bigelow Laboratory, and Northeastern’s Roux Institute. The AUSA Annual Meeting is in Washington October 12 to 14. MIT Lincoln Laboratory holds its ISR Systems and Technology Workshop in Lexington October 27 to 29, invitation only.

Portfolio Corner

Human role-players are the gold standard in training. They are also costly, slow to book, and hard to staff. So most people first face the hard conversation live. Delta AI, in Sandwich, Massachusetts, wants to fix that. It builds AI role-players for scenario training. Switch from a hostage negotiation to a tough command talk in one click, with a tutor and dashboards that flag where a trainee is weak. New North Ventures led its first round this year.

The role-play is the demo. The record is the product. Delta’s dashboards log how someone performed under pressure: what they decided, where they broke, whether they improved. That is the same attributable trail the deep dive says AI will owe about itself. CEO Mark Buonforte is a West Point graduate and Army veteran. Aristos Xanthus is his technical co-founder. The company is early and keeps its numbers private, so no customer count or contract figure here. Watch whether a tactical-dialogue tool becomes the system of record for readiness, for people first and machines next.

The Sovereignty Premium Crosses the Atlantic

Jeremy Hitchcock — Wed, 30 Sep 2026 10:31:44 GMT

For a decade, American institutional capital treated European defense as a discount bin. Fragmented buyers, small national budgets, no path to a real exit. This week the University of California’s investment office, which manages roughly $190 billion, wrote its first ever direct check into a European company. The company was a drone maker whose aircraft have logged more than 50,000 hours over Ukraine. Two days before that money closed, Britain’s new government stood at its party conference and called defense the engine of national reindustrialization. The discount is gone. What is replacing it is a premium, and this week it was paid in dollars.

The Deep Dive

A quintupled valuation, tested in a war zone

On September 23, TEKEVER closed the first tranche of a Series D at $580 million, led by UC Investments and Baillie Gifford. The round set the company’s post-money valuation at $6.4 billion, on a pre-money mark just under $6 billion. New strategic investor Merlyn Advisors joined, alongside returning backers Ventura Capital, Iberis Capital, and Crescent Cove. TEKEVER, founded in Portugal and now headquartered in Britain, builds unmanned aircraft, the software that flies them, and the intelligence services wrapped around them. CEO Ricardo Mendes has said the money funds manufacturing, acquisitions, and international expansion.

Trace the capital ladder and the repricing is stark. TEKEVER raised a 25 million euro Series A in late 2021 at a valuation near 100 million euros. It raised a 70 million euro Series B in October 2024 with Baillie Gifford and the NATO Innovation Fund. It raised an estimated 400 million pound round in May 2025 at roughly $1.3 billion. Sixteen months after that, the number is $6.4 billion, and the company has raised close to $1.2 billion in total. The ladder itself is the point: European sovereign-defense capital carried the early rounds, and now American institutional money is climbing on at the top. This is only a first close, so more is expected before the Series D is done.

What underwrites the curve is not a demo. TEKEVER’s systems have flown more than 50,000 operational hours over Ukraine since 2022, and Britain’s Ministry of Defence selected the company for its CORVUS surveillance program, a contract worth up to 400 million pounds over ten years. That is a ceiling on a framework, not money in the bank, but it is a procurement channel most defense startups never reach.

Why sovereignty became a valuation input

The old European discount rested on a real problem: no single buyer big enough to build a champion. That problem is inverting. Governments now want sovereign capability badly enough to say so out loud, and they are structuring budgets to prove it. On September 28, Britain’s finance minister John Healey told the Labour conference that a new age of industrialization would run through British shipyards and multi-billion pound defense projects, including three new floating docks for submarines on the Clyde. Prime Minister Andy Burnham has committed to the NATO target of 3.5 percent of GDP on core defense by 2035. Defense Secretary Wes Streeting put it plainly: back British by buying British.

The same pull showed up outside the capital markets this week. On September 25, Ukraine opened its combat AI training data to Britain through a first-of-its-kind partnership, letting a dozen British firms build on more than 5 million annotated battlefield frames. London described the goal in one phrase: sovereign capability at home. When a government treats another nation’s war data as a strategic asset to be secured and localized, the companies that can turn that data into deployed systems inherit the same premium the capital markets just handed TEKEVER. Battlefield validation is becoming a national resource, not just a corporate one.

The mechanism investors are pricing sits in the gap those statements create. Governments want domestic suppliers, but they cannot finance every one. Britain is already staring at a £4.7 billion hole in its own Defence Investment Plan. Sovereign demand is rising faster than sovereign money can fill it, and the space between the demand signal and the government checkbook is exactly where private capital earns its return. A company that is battlefield-proven and sits inside a national procurement pipeline is no longer a discount. It is scarce, and scarcity is what gets repriced.

Why the check came from California

The identity of the lead investor is the real signal. UC Investments had never made a direct investment in Europe. Its first one is a defense company. Baillie Gifford, already on the cap table since the Series B, doubled down. American institutions are crossing the Atlantic because the overlap they want, deployed capability plus sovereign demand plus AI autonomy, is priced more attractively in Europe than at home.

The home market pushes the same way from the other side. US defense venture is minting mega-rounds at the top, but the exit picture is bifurcated: venture dollars keep surging while strategic acquirers, the legacy primes, have already absorbed most of the transformational targets and are leaving a thinner middle. Europe offers a second venue that Washington’s exit math does not: national champions with real revenue, real deployment, and governments legally committed to buying from them. For a fund with a twenty-year horizon, sovereignty is not a political slogan. It is a durable source of demand that does not reprice with the news cycle, and it now comes with an institutional bid deep enough to underwrite a $6.4 billion valuation.

The Investment Takeaway

New North Ventures reads this as a repricing, not a bubble. The premium is accruing to a specific combination: a product deployed under real operating pressure, a domestic buyer that is compelled to purchase, and enough AI autonomy to scale without scaling headcount. Two of the three are not technology at all. They are a track record and a customer with a mandate. We think the same logic holds on this side of the ocean, and it favors companies wedged next to a guaranteed government buyer over those selling into a discretionary one. The founders who win the next cycle will be the ones who can point to hours flown, rounds fired, or missions run, and to a customer that has no sovereign alternative but to keep buying. Underwrite the mandate, not the demo.

The Funding Ledger

Six dual-use rounds from the past week, all disclosed between September 22 and 24:

TEKEVER raised $580 million in the first close of a Series D led by UC Investments and Baillie Gifford at a $6.4 billion valuation. Lisbon and London. Ukraine-proven unmanned aircraft and the intelligence layer around them.
Mesa Quantum raised nearly $12 million led by Playground Global. Boulder, Colorado. Chip-scale atomic clocks for timing and navigation when GPS is jammed or denied.
Pilgrim raised $25 million in seed funding at a reported $150 million valuation, led by Buckley Ventures. Redwood City, California. An airborne system, ARGUS, that samples the air and sequences it to catch biological threats early.
StandardX raised £10 million in seed funding led by Vsquared Ventures and East X Ventures. London. Accelerator-made medical isotopes now, tritium for fusion later.
Hughes Precision raised more than ₹250 crore, roughly $30 million, combining primary and secondary capital. Goa, India. Small and medium-caliber ammunition, and an approved supplier to the US Department of Defense.
TacnIQ secured the first half of a planned $3 million pre-seed from In Group Holdings. Singapore. Tactile-interaction data for robots, the one large training corpus the open internet never generated.

Set TEKEVER’s headline round aside and the other five share one bid: physical scarcity, not the model. American money bought independence from GPS and an early-warning layer for pathogens. British, Indian, and Singaporean money bought isotopes, ammunition, and the sense of touch. Each of these buys something a competitor cannot download: a production line, a calibrated sensor, a supply chain, a national-lab partnership. Notice too that two of the six are timing and navigation plays, and two are physical-supply plays for isotopes and ammunition. Capital is treating the boring, hard-to-copy layers, clocks and calibers, as the defensible ones. The model is the commodity. What sits around it is the asset.

The Northeast Desk

New Hampshire. BAE Systems launched a new family of electronic-warfare products on September 15, called Shadow EW, aimed at small airborne platforms. The design work and software are done at the company’s Nashua facilities, with production in Cedar Rapids, Iowa. The line runs on commercial microchips and open standards with field-upgradeable software, which is the point: BAE’s Rebecca Cruz framed it around countering threats that change faster than hardware cycles. The timing tracks the money. The Army requested $156 million for its Electronic Warfare Advanced Technology program in fiscal 2027, up 81 percent from the year before, and Nashua is where a chunk of that demand gets designed.

The regional throughline. New England is the closest thing the United States has to the sovereign-procurement moat that just repriced TEKEVER. The two-yard submarine base across Groton and Quonset Point, BAE’s electronic-warfare labs in Nashua, and MIT Lincoln Laboratory in Lexington are all suppliers a government cannot easily replace. The scale is real: Electric Boat alone is working to hire 8,000 people across its yards this year to meet the submarine build rate. When capital pays a premium for battlefield validation next to a captive government buyer, this is the region that already has both, and the workforce pipeline is the binding constraint on turning that advantage into ships and systems.

Watch the calendar. Maine Blue Economy Week runs in Portland from September 30 to October 2, convened by the Gulf of Maine Research Institute with the University of Maine, Bigelow Laboratory, and Northeastern’s Roux Institute. The AUSA Annual Meeting fills the Walter E. Washington Convention Center in Washington October 12 to 14. MIT Lincoln Laboratory holds its ISR Systems and Technology Workshop in Lexington October 27 to 29, invitation only.

Portfolio Corner

Human role-players are the gold standard in training. They are also expensive, slow to schedule, and hard to staff with the right background, which is why most soldiers, officers, and police rehearse the hardest conversations for the first time when they are live. Delta AI, based in Sandwich, Massachusetts, is trying to remove that constraint. The company builds AI role-players for scenario-based training, so a user can move from a hostage negotiation to a difficult command conversation at the click of a button, with a tutoring layer and dashboards that show where a trainee is weak. New North Ventures led the company’s first venture round earlier this year.

Delta is a clean expression of the week’s thesis. Capital just paid a premium for AI wrapped around a scarce human capability, and training is one of the scarcest: the skilled human counterpart who can play a hostile negotiator, a panicked civilian, or a difficult subordinate on demand. Delta is productizing that counterpart and selling it to both the private sector and the Department of Defense. Co-founder and CEO Mark Buonforte is a West Point graduate and Army veteran with more than a decade in and around the Department of Defense, and he brought in Aristos Xanthus as technical co-founder. The company is early and holds its numbers close, so we are not putting a customer count or a contract figure in print. What is worth watching is whether a tool that started with tactical dialogue becomes the system of record for how a force measures readiness, because the dashboards are quietly the more valuable half of the product.

Orbit Is No Longer a Parking Lot

Jeremy Hitchcock — Wed, 23 Sep 2026 10:30:54 GMT

Photo by Mara F on Unsplash

The two biggest moves in national security space this week were not launches. On September 18 the Pentagon picked Northrop Grumman and True Anomaly to build prototype satellites that can hunt other satellites in geosynchronous orbit. Two days earlier, Impulse Space raised $308 million to build vehicles that move payloads between orbits in hours. A contract and a check, same week, buying the same thing: the ability to move and to see in a crowded sky. Orbit used to be a place you launched into and parked. Now it is a place you maneuver through, and the companies that matter own the maneuvering and the watching, not the ride up.

The Deep Dive

The contract

The Pentagon's Defense Innovation Unit picked Northrop Grumman and the startup True Anomaly to each build a prototype for GHOST-R, satellites that inspect and track other objects in geosynchronous orbit, 22,000 miles up, where the military parks its most valuable communications and missile-warning craft. Northrop will use a commercial ESPASat-L bus and a commercial optical sensor, delivered in 24 months.

The buses are commercial. The Space Force is buying smallsat production lines and bolting on the sensors and software that make the mission classified, not commissioning bespoke decade-long designs. GHOST-R is a prototype tranche feeding RG-XX, a program to field a fleet of geosynchronous reconnaissance satellites by the end of the decade. The prototype is the audition; the fleet is the market.

Northrop is a prime that can manufacture at scale. True Anomaly is barely four years old, and its Jackal spacecraft is built for exactly this: maneuvering up to another satellite, imaging it, shadowing it. Pairing them on the same 24-month race is a hedge, the incumbent for production, the startup for reflexes, and it compresses the timeline that used to protect incumbents. A startup that can field a maneuvering inspector in two years is a second source, not a science project.

The capital

Impulse Space, in Redondo Beach, closed a $308 million extension to its Series D, taking the round to $808 million, backed by 137 Ventures, DFJ Growth, Lux Capital, and Valor Equity Partners. It builds neither satellites nor rockets, but the tugs and kick stages that move things once they are in space: Mira, a precision maneuvering vehicle already flown, and Helios, a stage that carries a payload from low orbit to geosynchronous orbit in under a day.

In July, the Space Force named Helios the first upper-stage vehicle ever made a prime in the National Security Space Launch program, historically reserved for rocket companies. Getting to orbit and moving once there are now two procurements, and the second is worth paying for alone. The $308 million says investors read it the same way and are funding capacity ahead of the contracts.

What changed

For thirty years, orbit was static. You launched a satellite, it held its slot, it stayed put and worked. The contest, when there was one, was who could launch more and build the most exquisite payload. Nobody was coming to bother your satellite once it arrived.

That is over. Geosynchronous orbit is congested and contested, with foreign craft making close approaches to inspect and shadow American assets. GHOST-R buys the ability to see what is maneuvering near your satellites; Impulse sells the ability to move your own faster than an adversary can react. The industry's shift to laser crosslinks, at 10 gigabits per second this week, hardens a third leg, a mesh harder to jam than radio.

A co-orbital threat is cheap to pose and expensive to ignore. An adversary can nudge a small craft toward your half-billion-dollar communications satellite for a fraction of its cost, and the mere possibility forces you to watch and be ready to move. Defense here is not a bigger satellite. It is faster information and faster maneuver, bought in quantity, so no single asset is a single point of failure. That is why the Space Force funds proliferated commercial-bus tranches instead of a few exquisite platforms, and why a company selling repositioning raised a third of a billion dollars in a week.

The commodity is the part everyone can now buy: commercial buses, commercial launch, even the sensors. The scarce parts are maneuver, the sensor tuned to the mission, and the software that turns a track into a tasking decision before the adversary acts on the same picture.

The Investment Takeaway. We believe the durable positions in national security space are awareness and navigation, not the bus or the booster. New North Ventures backs two. HawkEye 360 geolocates radio-frequency emitters from orbit, domain awareness pointed both at Earth and at other spacecraft, from a commodity constellation. Psionic builds navigation for when the space layer is jammed, the assured positioning a maneuvering force needs when it cannot trust the signal from above. Both sell the answer, not the hardware that carries it. As the buses commoditize, the companies that own the sensing and the decision loop keep the margin; the ones selling the ride up compete with catalogs.

The Funding Ledger

The week's dual-use checks, and what they bought:

Cornelis Networks: $205 million, networking. The Wayne, Pennsylvania company builds the interconnect that lashes AI and supercomputing clusters together, the plumbing behind national-lab and defense computing. Backed by IAG Capital Partners.
Arcee AI: $150 million Series B above a $1 billion valuation, led by Vista Equity Partners. The San Francisco company builds open-weight models a defense or intelligence customer can run behind an air gap.
Mazama Energy: $135 million Series B, led by Centaurus Capital and Doerr Capital. The Texas company, incubated by Khosla, drills superhot-rock geothermal at Newberry Volcano in Oregon, firm power for grids and bases that cannot run on intermittent supply.
EUCLYD: more than 200 million euros, Series A, co-led by Samsung with ex-ASML chief Peter Wennink as chairman. The Eindhoven company builds energy-efficient AI chips, a European bet against depending on American and Asian silicon.
MIND: $72 million Series B, led by Crosspoint Capital. The Seattle company, Israeli-founded, automates data-loss prevention, stopping sensitive files from walking out the door as employees feed them into AI tools.
Comp AI: $34 million Series A, led by Grand Ventures and Roo Capital. The Aventura, Florida company automates the SOC 2 and ISO audits a vendor must clear before a regulated or government buyer will sign.

Five are American; only EUCLYD, in the Netherlands, sits abroad. None is a weapons platform. The money went to the substrate under national security: compute, chips, power, models, data security, and the compliance gate. A quiet week for missiles and drones does not send capital out of the sector. It sends it one layer down, into the infrastructure every weapon program runs on.

The Northeast Desk

New Hampshire's stake in the orbital buildout runs through Durham. The University of New Hampshire's Space Science Center and Space Tech Hub license instrument technology to industry, and in July, at the Farnborough airshow, UNH signed a partnership called The Concord with Space Hub Yorkshire in the United Kingdom. Yorkshire's delegation comes to Durham for the Northeast STTAR Expo we flagged last week.

In Rhode Island, the Navy's oldest warfare center is rewriting how it buys. At SENEDIA's Defense Innovation Days in Newport, from August 31 to September 2, NUWC Division Newport technical director Marie Bussiere told contractors the command is moving to what she called radical collaboration: biweekly office hours with its engineers, a published Long Range Acquisition Forecast so companies can see contracts coming, and open roundtables instead of hierarchy. Scott Sideleau announced a new challenge, Narragansett Sentinel, to protect the subsea communications cables where military and commercial traffic mix. The center, under Capt. Kevin Behm, wants the industrial base building pieces of the problem, not waiting for a single prime.

Watch the calendar. Narragansett Sentinel opens with a classified industry day at Division Newport in late October, the first step toward the Navy's BlueTIDE exercise in 2027. The Northeast STTAR Expo runs October 28 and 29 at UNH in Durham, suppliers on day one and a student career day on day two.

Portfolio Corner: Pitchfire

Disclosure first: Jeremy Hitchcock, who writes this newsletter, co-founded Pitchfire, and New North Ventures is an investor. Read the rest with that in mind.

Every venture firm drowns in inbound. A partner gets hundreds of cold decks a quarter, most off-thesis, and the good ones vanish into the noise. The tax is human: associates reading pitches for hours to decline ninety percent of them, founders waiting weeks for a no. Pitchfire ingests the decks and emails, researches each company against public sources, and scores it against a firm's stated focus, so a partner sees a ranked queue instead of an inbox.

The bet is familiar: the model is a commodity, the value is the workflow around it. Pitchfire is not building a frontier model. It is trying to own a firm's intake loop, the judgments a fund makes on every deal, and compound them into a system that sharpens with use. It is live for early partners, including other funds testing it on their own deal flow. It is early, and we are not publishing metrics we cannot yet stand behind. Firms that turn inbound into structured, searchable judgment will out-decide the ones still reading decks one at a time.

The Weapon and the Shield Are the Same Model

Jeremy Hitchcock — Wed, 16 Sep 2026 10:30:46 GMT

National Security Agency headquarters at Fort Meade, Maryland, where the agency is reorganizing around five mission centers, including one for artificial intelligence. NSA photo, public domain.

In the span of one week, a small cell in northern Yemen used a commercial AI coding tool to work on missile guidance software. Cyber teams in Brussels used a commercial AI model to find flaws in the European Union’s own code. The National Security Agency started a 30-day clock to rebuild itself around artificial intelligence. And China’s spy chief wrote that cybersecurity has entered an era of “AI versus AI.” The weapon, the shield, and the institution all reached for the same technology at once. The barrier that kept sophisticated capability in the hands of sophisticated actors is coming down, and the companies that matter now are the ones that own what the model cannot.

The Deep Dive

The weapon

On September 10, Anthropic published its latest threat report, Detecting and countering misuse of AI. One case describes a cell operating from northern Yemen that used Claude Code as a stand-in for a missile engineering team. Anthropic did not name the group, though, as The War Zone noted, the Houthis are based primarily in northern Yemen.

The operators ran parallel sessions: one writing code, one researching, one reviewing the work of the first. Their stated goals included a multi-stage ballistic missile with a range above 2,000 kilometers. They worked through guidance and flight-control software, then test-fired a guided rocket. By Anthropic’s account the test appears to have failed, and within hours the actors were back asking the model why.

Anthropic banned the accounts and found no evidence the group fielded an operational weapon. By then the actors had already built an offline simulation toolkit that runs without Claude. Safeguards blocked many of their requests, but not all, and the operators split the work across sessions so no single conversation revealed their full intent. The Yemen cell was one of several conventional-weapons cases in the report, alongside cases tied to China and Russia.

This was not a chatbot answering one dangerous question. It was a distributed engineering program run by a handful of people, compressing design, testing, and failure analysis into a single loop. A safety system that reads requests one at a time cannot see a missile program assembled from innocuous pieces.

The shield

The same week, Politico reported that ENISA, the EU’s cybersecurity agency, used an advanced OpenAI model to review source code for an EU project. The model found four vulnerabilities. One was rated high risk and could have let an attacker hijack accounts. All four have been fixed. The EU only won access to the most capable US models in July, after months of asking. Within weeks, the payoff showed up in patched code.

Europe spent months arguing about digital sovereignty and its dependence on a handful of American labs. The first thing it did with that access was harden its own systems. The dependence Europe worried about is now also load-bearing in its defense.

The institution

General Joshua Rudd, who leads both the NSA and US Cyber Command, started a 30-day clock this month to recast the agency into five mission centers: China, cybersecurity, artificial intelligence, combat support, and global intelligence. Each center gets its own chief, including a new head of AI. The target is full operational capability by January. At his confirmation hearing before the Senate Intelligence Committee in January, Rudd said his “first priority is speed.”

The agency last attempted a restructuring of this scope roughly a decade ago, and current and former officials still regard that effort as a failure that added bureaucratic confusion. Officials told The Record that Rudd and Deputy Director Tim Kosiba know the rapid realignment will “break things.” They are moving anyway. When an institution built on process decides speed beats stability, it is telling you how fast it thinks the other side is moving.

AI versus AI

Beijing is reading the week the same way. Chen Yixin, China’s Minister of State Security, named Anthropic’s Claude Mythos and OpenAI’s GPT-5.5-Cyber as signs of a “disruptive upgrade” in cyber capability. Writing in the journal of the Cyberspace Administration of China, Chen said the field is entering a phase of “vulnerability industrialization, fully automated attack and defense, and AI versus AI.” He did not allege either model had been used against China. When Beijing’s spy chief and American security teams cite the same model names in the same week, nobody holds an edge from access alone.

What actually changed

The capability that let a small cell stand in for a missile engineering team let Brussels stand in for a red team. Offense and defense now draw from the same well. The advantage no longer comes from having the model, because everyone has the model. It comes from what surrounds it: proprietary data, evaluation, the operational workflow, and the speed to act on what the model surfaces before an adversary acts on the same insight. The NSA’s stated goal is speed, not raw capability, for exactly this reason.

The offline toolkit is the harder lesson. Once a capability is assembled, it no longer depends on the vendor’s safeguards. Banning the accounts closes the front door after the tools have left through the back. Guardrails at the model are necessary and insufficient. Durable control sits with whoever owns the data and the workflow, not the weights.

The Yemen case is also the tell on proliferation. China and Russia already had these capabilities. A small, resource-poor group climbing toward work that used to require a national program is new. AI does not only make the strong faster. It raises the floor under the weak, and it does that fastest for the actors with the least to lose and the least to deter them.

The Investment Takeaway. We believe the defensible layer in national security AI is not the model. It is the proprietary data loop and the operational workflow around it, the part that compounds over time and cannot be walked out the door in a single session. We back companies that own that loop and turn model output into action inside a real operational cycle, in cyber, intelligence, and autonomy. Two of our portfolio companies sit squarely on this thesis: Reality Defender in deepfake and synthetic-media detection, and CrunchAtlas in active cyber defense for critical infrastructure. The model is a commodity input.

It also sharpens what we will not pay for. A thin layer over someone else’s frontier model, with no proprietary data and no operational hook, is a feature waiting to be absorbed by the next model release. Value accrues to whoever turns the technology into an outcome faster than the other side, and can prove it in an environment where the other side is using the same tools.

The Funding Ledger

The week’s biggest dual-use checks, and what they bought:

Stoke Space: $1 billion Series E, co-led by Point72 Ventures and Spark Capital. The Kent, Washington company is building Nova, a fully and rapidly reusable rocket, and has now raised $2.3 billion in total.
Positron AI: $875 million across two tranches at a $5 billion valuation. NEA co-led both, joined by Andra Capital, Atreides, Valor Equity Partners, and SemiAnalysis Capital in the first and Jim Clark in the second. The Reno company builds inference systems on commodity memory, sidestepping the constrained supply chain for high-bandwidth memory.
Suniva: $835 million in debt and equity, with senior credit from Goldman Sachs Alternatives and I Squared Capital and equity from Electron Capital Partners, Orion Infrastructure Capital, and Rubric Capital Management. The Norcross, Georgia company is the largest and oldest US merchant maker of monocrystalline solar cells and is merging with Nasdaq-listed SUNation Energy.
Mach Industries: $600 million Series C extension at a $3.7 billion valuation, from investors including Ribbit Capital, Infinite Capital, Bedrock, and Sequoia. The Huntington Beach company builds strike drones, counter-drone systems, and propulsion, and doubled its valuation in three months.
The Exploration Company: $450 million Series C, co-led by Bessemer Venture Partners, Atomico, and EQT’s Scaleup Europe Fund. The European company is building the reusable Nyx capsule, and calls this the largest Series C ever announced by a European space company.
Poseidon Aerospace: $60 million Series A led by TQ Ventures, with JAWS, G Squared, and Hanwha Asset Management USA participating. The Alameda, California startup is building Egret, an uncrewed regional cargo aircraft aimed at contested logistics, with a first full-scale flight expected by year end.

Every dollar here went into physical things: rockets, a capsule, chips, solar cells, strike drones, and cargo aircraft. Not one software wrapper. The Exploration Company is the only non-American name, and it posted Europe’s largest space Series C, which says sovereignty is now a fundable thesis on both sides of the Atlantic. Suniva stands out for a different reason. A domestic solar cell maker raised most of its money as credit, because lenders now underwrite supply-chain independence like infrastructure. Capital is voting for the industrial base, not the app layer.

The Northeast Desk

New Hampshire’s defense community gathered this morning. BENS, Business Executives for National Security, convened its New Hampshire national security ecosystem breakfast in southern New Hampshire, and New North Ventures helped host. The goal is to put the founders, primes, universities, and state officials building the state’s defense economy in the same room.

The University of New Hampshire will host the first Northeast Space, Technology, Trade and Research Expo in Durham on October 28 and 29, with the state’s Department of Business and Economic Affairs and the NH Aerospace and Defense Consortium. According to BEA data cited by UNH, New Hampshire has about 300 aerospace and defense companies with 10,500 direct employees and $7.5 billion in output, 3.6 percent of state GDP and growing toward 4 percent. For a state our size, that is a core industry.

In Rhode Island, Havoc signed a licensed co-production partnership with Taiwan’s CSBC Corporation on September 10 to field autonomous surface vessels for Taiwan’s maritime security. CSBC builds the vessels. Havoc supplies the collaborative autonomy software. “Congress wrote co-production with Taiwan into law in the FY2026 defense authorization,” said CEO Paul Lwin, pointing to Section 1266, which names uncrewed systems. (Havoc is a New North Ventures portfolio company.)

On the calendar: the NDIA Undersea Warfare Fall Conference runs September 21 to 23 in Groton, Connecticut, at the center of the submarine industrial base. The Northeast STTAR Expo follows October 28 and 29 at UNH in Durham.

Portfolio Corner: Headlamp

Around 200,000 service members leave the US military every year. Almost 12,000 of them, across the Defense Department and the Coast Guard, went through SkillBridge in the first half of fiscal 2024, the program that lets troops train with a civilian employer during their final months in uniform. The gap between those two numbers is the market Headlamp is building for.

Headlamp, which New North Ventures led at the pre-seed, runs what founder and CEO Steve Chang calls SkillBridge-as-a-service. Chang is a retired Special Forces officer with an MBA from Wharton. The problem he is attacking is two-sided. Transitioning troops do not know which employers are ready for them, and employers do not know how to run a compliant fellowship. Headlamp sits in the middle, placing veterans into Headlamp Fellowships and handling the training and administration around them.

SkillBridge grew so fast that the Defense Department paused approvals of new employers in 2023 to catch up. Meanwhile the shipyards, factories, and defense startups featured every week in this newsletter all say the same thing about hiring: the people are the constraint. Headlamp turns a national obligation into a hiring channel. The company is still early and quiet on its own metrics.

The Military Became an Energy Developer

Jeremy Hitchcock — Wed, 09 Sep 2026 10:01:19 GMT

Soldiers of the 38th Air Defense Artillery Brigade emplace a Patriot radar at night, July 19, 2025. The Army used the image to announce its Janus microreactor awards. U.S. Army photo by Capt. Frank Spatt.

Two things happened in the past ten days that belong in the same sentence, though nobody filed them that way. Private investors put $26 million into software that finds hidden capacity inside the electric grid, and the industry logged a second straight record year of global nuclear generation. Neither reads as a defense story. Both are one now. Over the past two weeks the federal government stopped treating electricity as a utility problem and started treating it as a weapons program: reactors going onto Army bases, foreign-made hardware pulled out of the grid by emergency order. Capital is already moving to the same map. This is the clearest sign yet that energy has become the defense sector’s next procurement lane, and most investors are still pricing it as clean tech.

The Deep Dive

The buyer moved first

On August 26 the Army named five companies to build nuclear microreactors at five installations under its Janus program, through other transaction agreements worth up to a combined $2.2 billion, paid against milestones through fiscal 2031. Westinghouse goes to Fort Drum in New York. Antares Nuclear goes to Fort Bragg. BWXT goes to Fort Campbell. General Atomics goes to Fort Hood. Radiant goes to Fort Benning. The program launched in late 2025 and traces back to executive direction from the spring of that year, so it is not new; we flagged Janus in December, when the Army named nine candidate sites. What is new is that the Army stopped studying microreactors and started buying them, at five sites, on the same day.

For a decade, advanced nuclear had a demand problem, not a technology problem. The reactors worked on paper and the customers stayed hypothetical. A base commander who wants power that keeps running when the commercial grid goes dark is not a hypothetical customer. The Army has now signaled it will pay for resilience directly, install by install, rather than wait for utilities to build it.

The number that matters is eight

Westinghouse says its eVinci unit is designed to run for eight years or more without refueling. A base that can generate its own power for eight years is a base that cannot be coerced by taking down the local grid, cannot be held hostage by a fuel-logistics chain, and does not go dark when the transmission lines around it fail or get attacked. That is the whole point of putting a reactor behind the fence. It is not cheaper power. It is power an adversary cannot reach. For an investor, that reframes the entire underwriting question. The value is not levelized cost. The value is assured availability, and assured availability is something a defense customer will pay a premium for and write into a requirement. Every generation technology that can credibly promise it, from microreactors to long-duration storage to on-site fuel cells, inherits that premium.

The grid became a national emergency

The same day, August 26, the White House signed an order declaring a national emergency over the country’s bulk-power system and moving to block foreign-produced grid equipment: large transformers, inverters, circuit breakers, industrial control systems, and grid-scale batteries tied to a list of embargoed and adversary states. The stated reasons were cyber backdoors and supply-chain dependence. The government’s position is that hardware sitting inside the grid can carry digital backdoors, and that a transformer built by an adversary is a standing intrusion, not just a procurement risk. The Department of Energy has 120 days to write the rules.

One order puts generation onto defense sites the government controls. The other pulls untrusted hardware out of the civilian grid the government depends on. Both treat electrons as terrain. Both assume the grid is contested. Neither is a subsidy program dressed up in security language. They are security programs that happen to be about power.

Capital is following the electrons

Venture capital reached the same conclusion from the other direction. On September 1 Gridsight, a Sydney-based company whose software helps utilities find capacity that already exists on their networks instead of building new lines, raised $26 million in a Series B led by Insight Partners. The pitch is blunt: as much as three-quarters of grid capacity can sit unused because congestion moves by place and time, and AI data centers plus electrification are about to make every spare megawatt valuable. On the same day, an Indian maker of silicon carbide chargers and inverters raised $4 million with a stated order book five times its revenue run rate.

Add the record global nuclear generation the World Nuclear Association reported on Monday, 2,702 terawatt-hours in 2025 and a second consecutive record. The constraint on both defense and AI is now the same constraint: firm, secure, controllable power. The companies that generate it, the companies that squeeze more out of existing wires, and the companies that keep foreign code out of both are converging into one investable category.

Why this is not the last nuclear cycle

Skeptics have seen advanced nuclear hype before, and they are right to flinch. The difference this time is the buyer. In the last cycle the customer was a regulated utility weighing a twenty-year rate case. In this one the customer is a service branch with a mission requirement, an appropriation, and a base it fully controls. That shortens the path from prototype to revenue in a way clean-tech economics never could. It also changes who wins. The advantage moves to teams that can satisfy a defense customer’s timeline, security requirements, and site constraints, not to whoever has the lowest projected cost per kilowatt-hour in 2035.

The Investment Takeaway. We think energy resilience is the most underpriced dual-use category in the market, because the buyer just became a first mover instead of a laggard. That reprices the whole stack. Generation is investable, from microreactors to long-duration storage. The grid-software layer is investable, because utilities now have to expose capacity and secure it at once. And the hardware-trust layer, the transformers and controllers that the emergency order just turned into a compliance mandate, is investable on a clock the government is setting. New North Ventures is underwriting energy the way we underwrite defense autonomy: as a category where the customer is real, the money is appropriated, and the timeline is short. Disclosure: our founder is co-founder and chairman of StarCube, a microreactor company, so read our conviction here with that interest in view.

The Funding Ledger

A quiet week for headline size, a loud one for direction. Nearly every notable round on September 1 sat in physical infrastructure or the defense economics around it.

Gridsight, $26 million Series B, led by Insight Partners. Software that helps utilities unlock unused capacity on existing grids.
Incera Solution, about $10 million Series A, led by SV Investment. Precision steering optics for satellite laser communication and defense pointing systems.
Kepler Aerospace, $8 million seed, led by Blue Ashva Capital. A vertically integrated satellite-and-intelligence stack aimed at defense and security customers.
Airbility, roughly $5 million Series A, led by Sazze Partners. Interceptor drones built to defeat hostile drones at a fraction of a missile’s cost.
xorlab, 5 million euros Series A extension, led by Spicehaus Partners. Behavior-based email security pitched on European data sovereignty.
Zenergize, $4 million pre-Series A, led by Giraffe Studios. Silicon carbide EV chargers and solar inverters, with a stated order book five times its revenue run rate.

Read the pattern, not the deals. Four of these six make something physical: optics, satellites, interceptors, power hardware. The two software rounds sell scarcity of a different kind: one finds capacity inside existing wires, the other sells European data sovereignty. Separately, Smack Technologies, a US startup building edge AI for dismounted troops, including a wrist-worn display still in prototype, drew coverage this week for the $61 million Series B it closed in August, and Palantir CEO Alex Karp agreed to be first lead investor in a still-unnamed defense-tech company from Mykhailo Fedorov, Ukraine’s defense minister until July, on undisclosed terms. Money is moving toward companies that own a hard, defensible piece of the real world, and away from thin wrappers on someone else’s model. None of the six is American. The lead checks came from Insight Partners and Sazze Partners in the United States and Spicehaus in Zurich as often as from local funds, while the builders sit in Australia, South Korea, India, and Switzerland. That gap is its own investment thesis for anyone underwriting the domestic industrial base.

The Northeast Desk

New Hampshire. GKN Aerospace is expanding its North Charlestown plant, a $16 million project announced in July that adds around 57,000 square feet and triples the site’s footprint, with a dedicated production cell for compressor blades and vanes for current and future engine programs. GKN says it means new skilled jobs but has not put a number on them, and it is still a jobs story as much as a parts story, in a town most defense readers could not find on a map. On August 31 the Pentagon signed seven-year framework agreements with Lockheed Martin and General Dynamics’ Ordnance and Tactical Systems unit to triple Patriot PAC-3 MSE and quadruple THAAD production, and that demand runs straight down into small New Hampshire suppliers such as Manchester’s Vibrac Precision Test Systems, which we profiled last month.

Southern New England. The region’s marquee defense gathering, SENEDIA’s Defense Innovation Days, ran August 31 to September 2 in Newport, its twelfth year, with workforce development as the headline theme and more than 400 people in the room. General Dynamics Electric Boat, which already employs more than 27,000 people across Groton, Quonset Point, and New London, is targeting about 8,000 new hires this year to build Virginia and Columbia-class submarines under the $76.6 billion award it shares with HII Newport News, booked July 29 and covered here last month. On August 24 Electric Boat and the New England Institute of Technology said they will triple the Warwick, Rhode Island training center from 30,000 to 100,000 square feet, with $20 million from the Navy’s submarine industrial base program and $25 million from Electric Boat, to push roughly 3,000 hires a year through it. The submarine industrial base does not have a capital problem. It has a people problem, and it is trying to hire its way out of it at a scale that reshapes the labor market from Rhode Island to eastern Connecticut. The same squeeze runs north to Portsmouth Naval Shipyard in Kittery, Maine, where attack-submarine overhauls compete for the same welders and machinists. For a fund that spends its days on software valuations, the binding constraint in the region’s largest defense enterprise turns out to be skilled trades.

Watch the calendar. On Wednesday, September 16, BENS is convening leaders from across New Hampshire’s national security ecosystem, including industry, academia, and federal and state government, for a breakfast in southern New Hampshire on opportunities to strengthen the state’s role in America’s national security and economic competitiveness. New North Ventures is helping host, and jeremy@newnorthventures.com is the contact if you would like to be there. The NH Tech Alliance’s Cybersecurity Summit is Thursday, September 10 at Manchester Community College, its Innovation Summit is October 6 in Nashua, and the Defense Investment Forum runs October 27 and 28 in Middletown, Rhode Island, which is where the SENEDIA crowd reconvenes this fall.

Portfolio Corner: CrunchAtlas

The grid the government just declared a national emergency still has to be defended one network at a time. That is the business CrunchAtlas is in. The company builds AI-driven active cyber defense for critical infrastructure, the kind of monitor, detect, evaluate, and act loop that operators need once an intruder is already inside the fence. Its framing, drawn from founder Ben Faberlle’s appearance on our Securing Our Future podcast last fall, is that detection alone is no longer the job. Attackers adapt, so defenders have to respond at machine speed, not analyst speed.

The regulatory wind is at their back. Federal regulators approved a reliability standard last year requiring internal network security monitoring inside utility networks, with compliance deadlines in 2028 and 2030. Translated, the country’s utilities are being told to watch the traffic already moving inside their walls, not just guard the perimeter. That is precisely the visibility gap active defense is built to close, and this week’s emergency order only sharpens the point: if a transformer can carry a backdoor, the monitoring has to assume the adversary is already home.

CrunchAtlas is early and deliberately quiet, and we are going to respect that here rather than put numbers on the table the company has not disclosed. If energy is now defense infrastructure, then the software watching that infrastructure from the inside is a defense product, and the buyers, the utilities and the operators, are the same customers the rest of this issue is about. New North Ventures backed the company early for exactly that reason.

Defense Tech Learns to Borrow

Jeremy Hitchcock — Thu, 03 Sep 2026 13:08:51 GMT

Castelion’s Project Ranger production campus in Sandoval County, New Mexico. Photo: Castelion.

Two of the largest defense raises of the last few weeks arrived with something novel: a debt tranche. On August 27, REGENT Craft, the seaglider maker in North Kingstown, Rhode Island, closed a $240 million Series B split evenly between equity and debt, with the loan half coming from Erebor Bank. A week earlier, the hypersonic startup Castelion paired $800 million of equity with a $250 million revolving credit facility. For a decade, defense tech was a software story funded almost entirely with equity. The money now building the factories looks more like project finance. In the same fortnight, a sensing roll-up called Lyntris had to trim its IPO to get public. Capital is flooding the top and the bottom of this market and stepping over the middle.

The Deep Dive: The Debt Tranche Is the Tell

Start with what REGENT actually did on August 27. The company raised $240 million, described plainly in its own release as “an equal mix of equity and debt.” Mare Liberum and AE Ventures, the venture arm of AE Industrial Partners, co-led the equity. The debt came from Erebor Bank, the new, OCC-chartered lender co-founded by Palmer Luckey. The round brings REGENT to $340 million raised to date and lands right as the company moves from prototypes into production at its new 255,000-square-foot plant at Quonset.

Read the structure, not the headline. A software company does not borrow $120 million. It has no factory to secure the loan against, no equipment to repossess, no purchase orders a bank can underwrite. REGENT does. It has a building, tooling, a US Marine Corps contract for its Viceroy platform, and an order book booked years out. Those are collateral. Debt shows up exactly when a company crosses from selling code to building things, because atoms can be financed in ways that atoms-free businesses cannot.

Castelion tells the same story at a different scale. Its Series C, led by JPMorgan, Andreessen Horowitz and Carlyle at a $13 billion valuation, carried a $250 million committed revolving credit facility alongside the $800 million of equity. Castelion is not a slide deck. It is trying to mass-produce hypersonic weapons, which means production lines, test ranges, and long-lead materials. The presence of committed bank credit in a Series C for a missile company would have been unthinkable three years ago. It is now a signal of seriousness.

Who is lending, and why it matters

Erebor was built specifically to bank the industrial base. It was chartered to lend to defense, manufacturing and hard-technology companies, and its application to the OCC proposed capping loans at roughly half of deposits, a deliberately conservative posture for a bank underwriting hard assets. When a lender like that steps into a seaglider round, it is making a claim: American defense manufacturing has become creditworthy. Factories, contracts and backlog are now things a regulated bank will lend against, not just things a venture fund will bet on.

This is a maturation marker. Equity funded the experimental phase of defense tech, the years when nobody knew whether a startup could win a program of record or build at scale. Debt funds the industrial phase, when the questions shift from “will this work” to “how fast can you build it and how do we finance the plant.” The arrival of credit does not replace venture capital. It sits on top of it, and it only becomes available once the venture bets have been proven out. It also does something subtler. It lets a founder scale production without selling more of the company, which protects the equity that the earliest believers already own. A debt tranche is a vote that the collateral is good and the business will be around to repay it.

The middle is where it gets hard

Now hold that against Lyntris. The sensing company, assembled by Trive Capital from Accelint and Vitesse, priced its IPO on August 18 below its expected range, cut the deal to 17 million shares from 24 million, and opened about 11 percent below its IPO price on its first day of trading as “LYNX,” for a valuation near $1.8 billion on roughly $451 million of trailing revenue. This is a real defense business with real revenue, and the public market still made it work for the money.

Put the three side by side and the shape of the market appears. Enormous private rounds at the top, now large enough to carry their own debt. A busy seed and early-stage layer at the bottom. And a thin, unforgiving public window in the middle, where even a company with $451 million of revenue gets a haircut. This is the barbell we have written about before, and the debt tranche is the newest evidence for it. Money is happy to fund the giants and happy to fund the experiments. It is skittish about the companies stuck in between, the ones too big for a clean venture story and too small or too early for a warm public reception.

What the barbell rewards

The barbell also explains who the strategic investors are. Lockheed Martin Ventures sits in REGENT’s cap table alongside Founders Fund and Japan Airlines. The primes are increasingly acting like venture investors, taking early positions in the companies whose technology they will later buy or integrate, precisely as their own acquisition budgets tighten. When a company reaches the scale REGENT is now approaching, that early strategic relationship is what turns a demonstration into a contract, and a contract into the backlog that a bank will finance. The flywheel runs prime interest to venture equity to federal contract to bank debt to factory.

The Investment Takeaway

We read the arrival of debt as a bullish signal about the category and a clarifying one about stage. Debt is confirmation that defense hardware is real, that these companies have assets and revenue worth lending against, and that the industrial base is being rebuilt with private capital. But debt follows contracts and collateral, which means it rewards the late stage. It is not where a seed fund plays, and it is not what a seed fund provides.

For New North Ventures, we remain consistent about how we underwrite. We invest at the beginning, before there is a factory or a Marine Corps contract or a loan. The new question is whether a company is building toward atoms or software that will one day be collateral. How can the roadmap end in something a bank will finance: a production line, a fielded system, a backlog? The companies that clear that bar are the ones the barbell eventually pulls upward. We would rather own them at the seed, at a rational price, than chase them once the debt markets have already blessed them.

The Funding Ledger

A quieter back half of the fortnight for defense after Castelion’s billion-dollar week. Crunchbase called it a “sparser” lineup, and AI assistants retook the top of the table. The dual-use money that did move went to hardware and the infrastructure beneath it, and it spanned the full size range.

REGENT Craft, $240 million Series B (equal equity and debt), co-led by Mare Liberum and AE Ventures, debt from Erebor Bank. Seagliders for maritime defense and commercial routes. North Kingstown, Rhode Island.
Castelion, $800 million equity plus a $250 million credit facility, led by JPMorgan, Andreessen Horowitz and Carlyle at a $13 billion valuation. Low-cost hypersonic strike weapons. Torrance, California.
Muon Space, $250 million Series C, led by Eclipse. Satellite constellations for weather and intelligence, with a new San Jose line targeting up to 500 satellites a year.
Gatik, $200 million Series D, led by the Qatar Investment Authority and Koch Disruptive Technologies. Autonomous middle-mile trucking, a dual-use logistics play.
Emerald AI, $150 million Series A, led by Energize Capital and DCVC. Software that balances AI compute against available grid power, the energy layer under every defense data center.
Mara, $7 million pre-seed, led by Khosla Ventures with a16z Speedrun. Spike, a low-cost automated counter-drone system aimed at the cheap FPV threats reshaping the battlefield.

The pattern: after two weeks of megadeals, the biggest checks rotated back to AI, and defense money concentrated on things that get built. Counter-drone at $7 million and seagliders at $240 million are the same bet placed at opposite ends of the size curve. Investors want hardware and the picks and shovels around it, not more software.

The Northeast Desk

New Hampshire. Appledore Marine Engineering of Portsmouth won a firm-fixed-price contract for architect-engineering services inspecting Navy waterfront facilities in the Mid-Atlantic, an indefinite-delivery award from NAVFAC with a ceiling of up to $15 million and work running through August 2031. It is not a headline number, but it is exactly the kind of quiet, recurring federal work that keeps a small Seacoast engineering shop busy for five years.

Rhode Island. REGENT’s raise is also a jobs story. The company’s Quonset plant is expected to create 300 jobs, with the potential for up to 750 over a decade, and REGENT is eligible for up to $13 million under Rhode Island’s Qualified Jobs incentive if it hits its hiring targets. REGENT calls the $240 million round the largest venture raise in state history, and Providence Business News ran it as a record. A seaglider factory on Narragansett Bay, staffed by New Englanders, financed in part by a bank betting on American manufacturing, is about as clean a version of the dual-use thesis as the region produces.

The through-line for the Northeast is the one REGENT embodies: the maritime industrial base is where New England’s defense advantage compounds. The shipyards, the composites shops, the Navy relationships and the trained workforce are already here. Capital is now arriving to match them.

On the calendar. SENEDIA’s Defense Innovation Days ran this week at the Newport Marriott, August 31 through September 2, gathering national security leaders, industry and the next-generation workforce in the heart of southern New England’s submarine and undersea cluster. Watch the workforce sessions in particular. The hiring surge across the regional maritime base is the constraint that everything else now depends on.

Coming up. On Wednesday, September 16, BENS is convening leaders from across New Hampshire’s national security ecosystem, including industry, academia, and federal and state government, for a breakfast in southern New Hampshire on opportunities to strengthen the state’s role in America’s national security and economic competitiveness. The conversation will focus on defense innovation, supply chain resilience, workforce, advanced manufacturing, dual-use technology, and opportunities to grow New Hampshire’s defense industrial base. It is a good one for New Hampshire and northern New England, and New North Ventures is helping to host. Space is limited; contact jeremy@newnorthventures.com for details.

Portfolio Corner: Holocron

New North Ventures led the pre-seed round in Holocron Technologies, whose whole premise is that the best intelligence advantage now comes from public information, if you can read it fast enough.

Holocron, founded in 2022, builds an AI platform for what it calls global domain awareness in science and technology. It ingests open-source signals, research publications, patents, news, financial disclosures, and surfaces where foreign technology is moving before it becomes a classified surprise. The founding team is built for exactly this: Tristan Yang came out of US Special Operations Command, and Addam Jensen spent time at Intel and Sandia National Laboratories. They are people who have felt the cost of a technology surprise and set out to build the early-warning system for it.

The reason we backed it is the reason this newsletter exists. The center of gravity in intelligence is shifting toward open sources. A breakthrough in a Chinese lab, a quiet hire, a new supplier relationship, a patent filing, these are visible in public data long before they show up in a briefing. The hard part is not access. It is the machine reading and the pattern recognition at scale, which is precisely the problem Holocron is built to solve. It is the same discipline we practice every week here, turned into a product.

Holocron has kept a low public profile since the raise, and we are not going to invent milestones it has not announced. The question we are watching is conversion: whether a strong open-source analytics capability turns pilot interest inside the national security community into a durable program. The category is right and the timing is right. Execution is the variable.

Autonomy Goes Brownfield

Jeremy Hitchcock — Wed, 05 Aug 2026 19:39:36 GMT

Two defense companies made news this week, both selling something called autonomy, both chasing opposite futures. Agon launched out of London with $23 million to build synthetic battlefields where new autonomous weapons learn to fight before they ever meet a real enemy. A few time zones west, in San Carlos, California, a company kept pushing more software inside crewed aircraft that are already flying real missions today. One bet replaces the pilot. The other makes the pilot better. Venture capital is in love with the first story. The Pentagon is signing production contracts for the second. The distance between those two facts is the most useful thing we can tell you this week.

The Deep Dive: Autonomy Goes Brownfield

Two roads out of the same trade

Defense autonomy has split into two businesses that share a word and almost nothing else. The greenfield road builds new uncrewed platforms from a clean sheet: strike drones, autonomous fighters, uncrewed surface vessels, the whole robot armada. The brownfield road takes the aircraft, ships, and vehicles the military already owns and makes them smarter with software. Greenfield gets the magazine covers and the unicorn valuations. Brownfield gets to revenue on a fraction of the capital. This week sharpened the contrast.

Why the installed base is the shortcut

Start with the physics of the defense budget. The United States operates thousands of crewed aircraft that will still be flying in 2040. Replacing them with autonomous platforms means clearing airworthiness, building new hardware, training a force, and winning a program of record from zero. Every one of those steps takes years and invites cancellation.

Retrofitting intelligence into an existing airframe skips most of that. The useful frame here is borrowed from self-driving cars, though aviation has no formal equivalent of the scale. Level 1 is today’s autopilot and autothrottle. Level 4 and Level 5 remove the pilot entirely, which is where the hard problems live: data maturity, heavy hardware, recertification, and a decade of program risk. The near-term money is in Level 2 and Level 3, the advisory band where software helps the crew manage complexity without triggering a full recertification of the aircraft. If the system rides on data, sensors, and a pilot interface rather than a heavy hardware install, it can field in months instead of decades. That is the brownfield bet: the fastest path to deployed autonomy runs through the fleet that is already paid for.

The government is voting with production clauses

Watch what the buyer signs, not what the market celebrates. Our portfolio company Beacon AI signed a four-year prototype agreement with US Special Operations Command this spring worth up to $49.5 million, and the agreement carries a production clause meant to move the software toward operational fielding across the 350-plus aircraft that SOCOM and Air Force Special Operations Command fly. Count the mobility work Beacon runs separately and the addressable fleet passes 1,400. That is a buyer paying real money to make crewed planes smarter, not to replace them.

Greenfield is getting production checks too, and the thesis has to survive that. In June the Air Force put both Collaborative Combat Aircraft into production, General Atomics’ FQ-42A and Anduril’s FQ-44A, four months ahead of schedule and against a target of more than 150 airframes by the end of the decade, with a separate six-year competition running for the autonomy software that will fly them. The first FQ-44A came off Anduril’s Ohio line in late July. So the question is not whether a clean sheet can reach production. It is what each dollar has to clear to get there. Beacon needed a prototype agreement and a software release. CCA needed two new airframes, a live-fire campaign, a billion-dollar factory, and the better part of a decade.

The pattern showed up again in the week’s biggest defense dollars, and none of them came from venture funds. The Navy awarded General Dynamics Electric Boat and HII a combined $76.6 billion for nine more Virginia-class and five more Columbia-class submarines, which Representative Joe Courtney called the biggest shipbuilding award in modern American history. The Space Force put $615 million into a second round of space-based airborne moving target indicator work, $397 million of it to Rocket Lab, with Systems and Technology Research and one undisclosed vendor splitting the rest. Note what that second award is. Not the constellation, which SpaceX won in May at up to $4.16 billion, but a deliberate hedge against depending on a single technical solution. One check buys more of a proven platform. The other buys insurance on a new one. In both cases the government, not the private market, is absorbing the risk of scale.

The commercial tell

There is one more reason to favor the retrofit layer, and it does not appear on any Pentagon slide. Brownfield autonomy has a civilian buyer. Beacon says it already works with several commercial airlines, and the same software that helps a special operations crew manage a long, high-workload sortie helps a regional carrier fly safer and burn less fuel. Greenfield military drones have a far thinner second market. There is a civil uncrewed industry in delivery, cargo, agriculture, and offshore survey, but it is small, slow, and buys nothing that resembles a strike drone. A platform that assists the pilot can amortize its research across airlines and militaries at once, which lowers the cost of capital and widens the moat. Dual-use is not a slogan here. It is a structural advantage that the brownfield side of the trade owns far more cleanly.

The counterargument, and why it is weaker than it looks

The greenfield bulls have a fair rebuttal. The future is uncrewed, they say, and retrofits are a bridge that gets torn down once the autonomous platforms mature. Two things blunt that. First, bridges collect tolls for a long time, and the crewed installed base is measured in thousands of airframes with service lives running past 2040. Second, the better objection is defensibility: is brownfield software just a feature that a prime can absorb? The answer turns on data. A platform that spreads from hundreds of aircraft toward a four-figure fleet accumulates high-context flight data that competitors cannot easily replicate, and that data compounds into the product. The open question is how much of it a vendor gets to keep, since the government holds broad rights to what its own aircraft generate. Answer that one well and this is the difference between a feature and a moat.

The Investment Takeaway

We think the near-term returns in defense autonomy are in brownfield, not greenfield. New North Ventures backs both sides of this market, but the timing math favors the retrofit. Software that rides the installed base can reach revenue and a production clause on a fraction of the capital and calendar a clean-sheet platform needs to get to the same place. The production clause is the tell. It is the buyer signaling that a capability is real enough to field, and the cheapest ones to earn are landing on the companies making the existing fleet smarter. When you underwrite an autonomy deal, ask one question first. Does this need a new platform and a new program of record to earn a dollar, or can it ride something already flying? Right now the second answer is worth a premium.

The Funding Ledger

Dual-use and defense-relevant rounds that crossed the wire the week of July 28 to August 3.

Agon (United Kingdom): $30 million total, a $7 million pre-seed led by Lakestar plus a $23 million seed led by XYZ Venture Capital and Lux Capital with Northzone. Synthetic battlefields to train defense AI, including counter-drone work against swarms. London, with a second office in Berlin. The week’s only pure-play defense round.
K2 Space (United States): $500 million Series D at a $6.8 billion valuation, led by Kleiner Perkins and ICONIQ, with Lightspeed, CapitalG, Altimeter, Spark, Sands, ARK, and T. Rowe Price accounts participating. Large high-power satellites, now past $1 billion raised. The Space Force is a named customer, and K2 has Golden Dome work with Anduril and a supply commitment under the Protected Tactical SATCOM-Global program through SES Space and Defense.
Multiverse Computing (Spain): a Series C targeting up to $570 million at a $1.7 billion pre-money valuation, co-led by Forgepoint Capital International, BNP Paribas’ Solar Impulse Venture Fund, and Bullhound Capital. Quantum-inspired compression that shrinks AI models onto smaller hardware, an edge and dual-use story. Announced July 27, just before this window, and the round is still open to additional strategic investors.
Onyx (Israel and United States): $113 million Series B at a reported $640 million valuation, led by Bessemer with TCV, Cyberstarts, Conviction, FirstMark, Vintage, QuantumLight, and G Squared. Onyx Security runs a control plane to discover, monitor, and govern enterprise AI agents. New York and Tel Aviv.
Act Security (Israel): $60 million across two rounds announced together, a $20 million seed led by Team8 and Bessemer and a $40 million Series A led by Notable Capital with Startpoint Capital and SVCI. Action-centric cloud access control spanning humans, workloads, and AI agents. Founded by the team behind Medigate.
Hush Security (Israel): $30 million Series A, with Akamai joining as a strategic investor alongside existing backers Battery Ventures and YL Ventures. $41 million raised to date. Identity and access control for AI agents and other non-human identities. Founded by the team that sold Meta Networks to Proofpoint.
ZuriQ (Switzerland): $25.5 million seed, led by Quantonation with Forward.one, Extantia, and Firgun Ventures. Trapped-ion quantum processors on a 2D Penning micro-trap architecture. An ETH Zurich spinout.

Read the pattern, not the list. One pure-play defense company raised this week, and it raised $23 million. The week’s biggest round, half a billion dollars into K2 Space, went to a dual-use satellite company with real Space Force and missile-defense exposure, which is the more interesting fact. Everything else went to the agent-security layer and to deep-tech infrastructure. Private capital is funding the substrate that autonomous systems will run on, while the government writes the checks for the platforms themselves. Procurement budgets always dwarf venture rounds, so the size gap between the Navy and Sand Hill Road is not the signal. The signal is which layer venture will underwrite without a program of record behind it.

The Northeast Desk

New Hampshire first. In Manchester, Vibrac Precision Test Systems is a 25-person shop whose calibration systems qualify the ball bearings that go inside Patriot missiles. Lockheed Martin is scaling PAC-3 MSE output from roughly 600 a year toward 2,000 over seven years, and Vibrac’s chief executive says he expects orders north of $100 million over the next four years. That is a small Granite State supplier riding a very large restocking cycle, and it is exactly the kind of company that never makes a national defense-tech list.

New Hampshire’s aerospace sector also had a moment abroad. Senator Jeanne Shaheen co-led a bipartisan congressional delegation to the Farnborough Air Show in July with Senator Jerry Moran, and championed a record New Hampshire industry contingent there, where aerospace stood out as the state’s top export sector at $1.7 billion in exports and 24,000 jobs in 2025. SENEDIA used the show to announce the Senator Jeanne Shaheen Aerospace and Defense Scholarship for young women entering the field. Retail politics and industrial policy in the same handshake.

Down the coast, the submarine industrial base got the check it had waited nearly three years for. The Navy awarded General Dynamics Electric Boat in Groton, Connecticut, along with HII’s Newport News Shipbuilding, a combined $76.6 billion for fourteen new submarines, work that will drive thousands of hires across southern New England through the end of the decade. Every hull pulls on a supplier web that runs up through Rhode Island, Massachusetts, and into New Hampshire and Maine machine shops. The region has been standing up training pipelines from Connecticut community colleges to the yards for years, and the award finally gives those hires a decade of visible demand to plan against. If you make pressure-hull hardware or precision components in this region, your order book just changed.

Mark the calendar. SENEDIA’s Defense Innovation Days, the twelfth annual, returns to the Newport Marriott in Rhode Island from August 31 to September 2. It is the one room where New England primes, startups, and Navy program offices are all in the same place.

Portfolio Corner: Beacon AI

Beacon AI is a New North Ventures portfolio company, and it is the cleanest example we have of the brownfield thesis in practice. The Silicon Valley company builds a software-first, hardware-light platform that augments pilots rather than replacing them. Its two products, Murdock and Lighthouse, sit on top of an aviation data platform that folds together aircraft data, weather, routing, and pilot actions to do three things: assist the crew in the moment, route around threats and weather, and monitor aircrew readiness on long missions.

The number that matters is the contract. In April, Beacon signed what it says is its thirteenth Department of Defense agreement, a four-year prototype OTA with US Special Operations Command worth up to $49.5 million, with participation from Air Force Special Operations Command and a production clause built in. The company frames its work as Level 2 and Level 3 pilot assistance, its own borrowing of the self-driving scale rather than any FAA standard, a deliberate step beyond autopilot and a deliberate step short of removing the pilot, precisely because that is the band where you can field useful autonomy without triggering a full aircraft recertification. Chief executive Matt Cox put the pitch plainly: “We build systems that help crews avoid unsafe actions, improve performance, and execute complex missions more effectively.”

The strategic logic is the data. Beacon’s platform is designed to spread across a special operations fleet of more than 350 aircraft, and a mobility fleet of more than a thousand beyond that, and every aircraft it touches produces high-context flight data that improves the product and is hard for a competitor to replicate. That is the difference between a feature and a moat, provided the data rights land the right way. We will be watching for the first production decision under the SOCOM agreement and for the platform’s move toward a formal program of record.

The Government Became the Growth Round

Jeremy Hitchcock — Wed, 29 Jul 2026 12:04:12 GMT

There is a company in Exeter, New Hampshire that has raised $117 million in venture capital across the life of its Series B. In the last eight weeks, the federal government has committed $566 million to it.

That ratio is the story. Not the rare earths, not the tailings, not even the China dependency that makes the investment urgent. The ratio. Federal capital is now arriving in quantities that complement private markets, in instruments that look less like grants and more like the growth rounds a crossover fund used to write. It is happening at the exact moment the traditional exit path for defense companies is getting narrower. Founders and investors are going to have to decide what that means.

What Phoenix Tailings actually got

On June 16, the Office of Strategic Capital at the Department of War signed a conditional commitment for a $500 million long-term loan to Phoenix Tailings. Two weeks earlier, the Department of Energy awarded the company $66 million under its Rare Earth Elements Demonstration Facility Program, against a $148 million total project cost with $82 million in non-federal cost share.

Phoenix processes mining waste or tailings into rare earth metals. Its Exeter facility opened last October, runs with about 30 people, and produces neodymium-praseodymium and dysprosium-iron alloys at an initial 200 tons per year with a stated path to more than 1,000. The OSC loan funds expansion at the existing Massachusetts and New Hampshire sites plus construction of a new separation and metallization plant the company calls the Freedom Facility. The site has not been announced.

Two things about that loan deserve emphasis, because most coverage skipped them. It is conditional, meaning nothing has been disbursed and financial close depends on due diligence not yet complete. And it is debt, not equity. Nobody took a board seat. Nobody took a preference. The cap table did not move.

Compare that to what Phoenix raised privately: a Series B that closed in three tranches, $43 million in December 2024, $33 million more in April 2025, and a $40 million amplification in February 2026 that was itself $30 million of equity plus $10 million of venture debt.

The federal commitments are roughly five times the venture capital, and they cost the founders nothing in ownership.

Why the money is there

This is the Federal Government investing in components and capabilities. It is that the IEA puts China at roughly 91% of global refined rare earth output and 94% of sintered permanent magnet manufacturing, and Beijing has spent fifteen months demonstrating what it intends to do with that position. Export licensing controls landed on seven medium and heavy rare earths in April 2025, expanded to processing equipment in October, and reached extraterritorially to foreign-made products containing Chinese-sourced material on December 1.

The affected elements are not exotic. Dysprosium and terbium are what keep a neodymium magnet from demagnetizing when it gets hot, which is to say they are what make a missile actuator work. An F-35 carries about 920 pounds of rare earth material. An Arleigh Burke destroyer carries 5,200. A Virginia-class submarine carries 9,200.

OSC Director David Lorch was direct about where the gap is: rare earth midstream processing capabilities are, in his words, “key shortage areas that need to be rapidly addressed.” Midstream, not mining. Separation and metallization. There are two or three facilities outside China that can do it. One of them is in Exeter.

The same pattern, at the other end of the check size

Phoenix is the large version. The Defense Innovation Unit is running the small version, and the mechanics are worth understanding because they are showing up in more and more pitch decks.

DIU has leaned hard into prize challenges under 10 U.S.C. 4025. The Mine Countermeasure Modernization challenge carries a $10 million purse. The Autonomous Vehicle Orchestrator challenge carries a ceiling up to $100 million. Specular MIST, which closed its first phase today, carries $5 million for maritime electronic warfare. Ground-Based Affordable Mass closes August 5 and sits inside a $250 million program budget.

For scale: as of last September, the largest prize purse in the history of the Department of Defense was $1.85 million. Federal prize spending across the whole government was $247,000 in FY2011. Something genuinely changed.

But look at what winning actually pays. Five companies won the Drone Dominance Lethality Prize Challenge in May, including Bravo Ordnance and Northrop Grumman. The cash prize was $10,000. What mattered was a government-funded safety review worth roughly $500,000 per vendor, preferred-provider status, and a pathway into a Phase 2 procurement of 60,000 units.

Bravo’s Kevin Landtroop said the thing out loud: winning “absolutely changed the caliber of discussion we’re having with investors, suppliers, other customers/partners.”

That is not financing. That is customer-validated due diligence, performed in public, for free. Which is arguably worth more than the check.

The discipline this requires

This additional source of capital to founders is an amazing accelerant. Founders should explore this financing route but be eyes wide open about several factors.

Federal capital is politically volatile in a way venture capital is not. DIU’s appropriation went from $87 million in FY2022 to $983 million in FY2024, a 431% increase. Enacted FY2026 came in at $430 million, less than half the peak and roughly half what the House itself passed. Announced challenge ceilings are rising on a shrinking base. A company that underwrote to the FY2024 trajectory is in trouble.

Announcements are not obligations. The Phoenix loan is conditional. DIU’s own Specular MIST solicitation states plainly that “there is no guarantee that follow-on awards or contracts will be awarded.” GAO found DIU awarded 450 prototype agreements from FY2016 to FY2023 with about $1.7 billion obligated, and 51% of completed prototypes transitioned to production. That is a good number by government standards. It is a coin flip by underwriting standards.

The exit math got worse while the funding math got better. US defense tech M&A ran 18 deals worth $4.6 billion in 2025. Through May 22 of this year it ran 14 deals worth $1.2 billion, per White & Case. Deal count held. Average deal size fell from roughly $256 million to $86 million. Meanwhile venture funding into the sector hit $14.6 billion across 107 rounds through May, against $9.6 billion and 206 deals for all of 2025. More money in, less value out the middle. The primes are writing venture checks instead of acquisition checks.

The Investment Takeaway

Non-dilutive federal capital is the most underpriced input in defense tech right now, and it is being systematically misread in both directions.

Founders who chase it as a substitute for revenue end up building companies whose only customer is an appropriations cycle. Investors who see a press release about a $250 million challenge and mark it as $250 million of pipeline are marking a lottery ticket.

The correct read is narrower and more useful. Federal capital is an accelerant on top of a commercial thesis, and its real value is signal, not cash. Phoenix Tailings did not get $566 million because Washington liked the pitch. It got it because it built an operating plant in New Hampshire that makes metal the country cannot otherwise buy, and the government confirmed that by underwriting the next one.

We underwrite to the commercial business. We treat the federal commitment as third-party diligence we did not have to pay for. And we want to be early enough that when the government does show up, it is a step-up rather than a rescue.

The Funding Ledger

Dual-use rounds announced the week of July 20.

Antares, $470 million Series C led by Paradigm and Caffeinated Capital, structured as $370 million equity plus $100 million debt. Compact nuclear microreactors for US military installations, targeting 2028 deployment.
Cathedral, $160 million at a $1.4 billion post-money valuation, led by Andreessen Horowitz and Sequoia. AI-driven military cyber operations, founded by four former DOGE staffers.
Enigma, $71 million seed led by Index Ventures and Ribbit Capital. Robot-agnostic AI models and interfaces, with more than 100 robots in the field collecting interface data.
Istari Digital, $22.6 million. Digital engineering software for military design workflows.
Nuclear Turbines, 15 million pounds led by IQ Capital. A BAE Systems spinout applying jet turbine technology to compact reactors.
ASIO Technologies, $15 million first institutional round led by Protego Ventures. Tactical command and control plus jam-resistant navigation for GPS-denied environments.
1872, $15 million seed led by The O.H.I.O. Fund. Autonomous steel fabrication with coordinated robotics.

Two of the three largest rounds this week were nuclear. That is not a coincidence and it is not really an energy trade. Antares is selling power to installations, and Nuclear Turbines is a defense prime spinout. Compute demand and base resilience are converging on the same hardware, and the capital is following the convergence rather than either sector on its own. New North Ventures’ StarCube continues to execute as well.

One note on method: a $180 million Series A for the endpoint security company Glow was widely reported this week, but Crunchbase’s own roundup lists the same round with the same lead at $100 million. We are holding it until the discrepancy resolves rather than printing a number we cannot stand behind.

For context on the week before: Helsing raised $1.8 billion at an $18 billion post-money valuation on July 13, led by Lightspeed and General Catalyst. It is the largest defense startup round in European history, and the company’s emphasis that it remains “predominantly European-owned” is doing real strategic work.

The Northeast Desk

Hooksett, NH. Marmon Defense broke ground on a $40 million, 350,000-square-foot manufacturing plant on July 16, more than four times its current footprint. The plant makes naval cables and wiring for Navy shipbuilders in Mississippi and Virginia. Roughly 80 jobs next year, 400 over five years, completion targeted June 2027.

Londonderry, NH. Envision Technology cut the ribbon on July 7 for the Village at Technology Park, a 110-acre development anchored by a 234,000-square-foot headquarters. Envision builds electro-optical and infrared systems and holds two five-year DoD contracts totaling $350 million. Expected 1,500 permanent jobs. Founder Ken Solinsky previously built Insight Technology and sold it to L-3 Harris. The development includes 440 apartments and a 200-child childcare center, which is the most honest answer anyone in New England has given to the defense workforce problem.

Andover, MA. Raytheon won a $1.8 billion SPY-6 radar production and sustainment contract on July 21, with options that could take it to $3.3 billion. RTX has put more than $800 million into modernizing radar manufacturing and plans to double SPY-6 output by 2028.

Maine. The state’s Forest Bioproducts Advanced Manufacturing hub won $20 million on July 27, the only New England award in a $169 million, six-hub round that Commerce rewrote around national security criteria. Worth reading as a signal about what federal innovation money now rewards.

Worth knowing. The FY27 NDAA includes a CMMC assessment grant program offering up to $100,000 per small business, with roughly 840 New Hampshire firms eligible. If compliance cost is what is keeping you out of the defense supply chain, that is the door.

On the calendar. BlueTIDE 2026, the maritime and dual-use demonstration hosted by Polaris Tech Bridge with NUWC, runs August 26 and 27 in Newport. SENEDIA Defense Innovation Days follows August 31 through September 2, also in Newport. The NH Tech Alliance Innovation Summit is October 6, with its Speed Venture Summit on October 13.

Portfolio Corner: Auriga Space

Auriga builds electromagnetic launch systems. The long-term goal is a multi-kilometer accelerator that puts small satellites into orbit without a rocket. That is a decade-scale ambition, and the interesting thing about the company right now is what it is doing with the hardware in the meantime.

On July 1, Auriga opened the first commercial hypersonic weather-effects testing service, using its Prometheus magnetically levitated linear accelerator. The rig positions water droplets or particles in a test article’s flight path, drives the article through them at flight-representative speeds, and recovers it for analysis. Dozens of tests in a single week under repeatable conditions. Axiom Materials is the pilot customer, validating oxide-oxide ceramic matrix composites. Prometheus is already operating under Air Force and Missile Defense Agency contracts, and Auriga is now selling time on it commercially.

Two weeks later, on July 15, the company signed a three-year CRADA with the Army’s DEVCOM Armaments Center to evaluate electromagnetic launch as a counter-drone interceptor. The economics are the entire argument. An APKWS shot runs $25,000 to $40,000. A Stinger runs $430,000 to $480,000. A Patriot PAC-3 runs $4.2 million. Against mass drone attack, those numbers lose on their own. An electrically accelerated projectile carrying no propellant resets the cost curve, and the magazine depth is a function of how much power you can put behind it.

New North Ventures has backed Auriga since the seed. Founder and CEO Winnie Lai was profiled in Forbes on July 10.

Also in the portfolio: SensusQ, the Estonian intelligence-fusion company, was acquired by German drone maker Quantum Systems on July 10 as part of Quantum’s expansion into Tallinn. Terms were not disclosed. Congratulations to Marko Kaseleht, Villiko Nurmoja, and the team.

The Race to Secure Critical Infrastructure: From Data Centers to Shipping Lanes

Wed, 22 Jul 2026 10:03:49 GMT

The Gulf’s real front line in the Iran conflict runs through its water, cloud, and cables—and the links between them

The conflict in the Gulf is no longer defined solely by missiles, oil tankers, or the Strait of Hormuz. According to a recent Atlantic Council analysis, the region’s true front line now runs through interconnected critical infrastructure including water, cloud computing, undersea communications cables, and the digital systems that link them. Recent attacks demonstrated how modern conflicts can disrupt entire economies without directly targeting military assets, exposing vulnerabilities that extend far beyond the battlefield.

Hybrid warfare is increasingly focused on civilian infrastructure. GPS spoofing disrupted the navigation of more than 1,100 vessels in a single day, cyberattacks against cloud infrastructure spilled over into banking and payment systems, and strikes on desalination plants threatened the primary source of freshwater for Gulf nations. At the same time, cyberattacks across the UAE surged dramatically, illustrating how physical and digital attacks are now executed in tandem to maximize economic and societal disruption.

Another growing concern is the Gulf’s role as a global digital chokepoint. Beyond transporting a significant share of the world’s energy, the region is also home to critical undersea fiber optic cables that carry international internet traffic and connect massive AI data centers across Europe, Asia, and the Middle East. The conflict has already delayed major cable deployment projects, underscoring how adversaries can threaten global connectivity without ever severing a cable.

Future resilience will require governments to think beyond protecting individual assets. Water systems, energy infrastructure, cloud providers, financial networks, communications cables, and cyber defenses must be secured as one integrated architecture because attacks on any single node can quickly cascade across the entire system. For national security leaders and investors, the lesson is clear: tomorrow’s strategic competition will increasingly be fought through critical infrastructure and the digital networks that underpin modern economies.

US and Allies Warn Russian Hackers Are Targeting Critical Infrastructure Routers

A new joint cybersecurity advisory from the United States and 12 allied nations warns that Russian state backed hackers are actively targeting vulnerable routers and networking devices to gain access to critical infrastructure worldwide. The campaign, attributed to Russia’s Federal Security Service (FSB), focuses on exploiting weak passwords, outdated software, and poorly configured edge devices. The goal is to quietly infiltrate sectors including energy, communications, healthcare, transportation, water, and defense. Once inside, attackers steal network configurations and credentials that can enable long term espionage or future disruptive cyber operations.

The advisory urges organizations to treat routers as critical security assets rather than routine networking equipment. Recommended defenses include replacing default credentials, patching known vulnerabilities, disabling unnecessary services, implementing multi-factor authentication, and closely monitoring network traffic for suspicious activity. The warning underscores a broader trend in modern cyber warfare: adversaries are increasingly targeting overlooked infrastructure components to establish persistent access to systems that underpin national security and the global economy.

Opening Remarks: Congressional Wargame on AI-Enabled Cyber Threats to U.S. Critical Infrastructure

At a recent CSIS Congressional War Game, lawmakers and national security experts examined how artificial intelligence is reshaping the cyber threat landscape facing U.S. critical infrastructure. The exercise explored a scenario in which China has highly advanced AI capabilities, enabling faster, more sophisticated cyberattacks against essential sectors such as energy, communications, transportation, and other critical infrastructure. While the opening remarks were public, the war game itself was conducted behind closed doors with members of the House Homeland Security Committee and the House Select Committee on China to assess potential vulnerabilities and policy responses.

The event underscores a growing consensus in Washington that AI is transforming cybersecurity into a strategic competition between nations. AI can dramatically accelerate the speed, scale, and sophistication of cyber operations, allowing adversaries to identify vulnerabilities, automate attacks, and coordinate disruptions across interconnected systems. By convening policymakers in a realistic war game environment, CSIS aims to help Congress better understand the risks posed by AI-enabled cyber warfare and inform future investments in cyber resilience, critical infrastructure protection, and national security preparedness.

More links to explore:

HawkEye 360 Details Tactical Direct Downlink in Valiant Shield Exercise

New North Ventures portfolio company HawkEye 360 demonstrated tactical direct downlink capabilities during the multilateral Valiant Shield military training exercise, working alongside Lockheed Martin to establish a machine-to-machine data pipeline. The demonstration showcased HawkEye 360's ability to deliver signals intelligence directly to operational units in near real time, bypassing traditional ground station relay architectures that introduce latency and operational complexity. This capability is critical for time sensitive targeting and situational awareness in contested environments where adversaries can relocate or adapt faster than traditional intelligence dissemination cycles allow.

The successful integration with Lockheed Martin systems during a large scale military exercise signals growing demand for commercially provided space based RF intelligence that can be operationalized at tactical edges.

Thanks for reading Securing Our Future! Subscribe for free to receive new posts and support our work.

Subscribe now

Securing Digital Trust: The Fight Against Deepfakes & Disinformation

Wed, 15 Jul 2026 10:00:44 GMT

Before diving in, it's worth distinguishing between two terms that are often used interchangeably. Misinformation is false information shared without the intent to deceive, while disinformation is intentionally created or spread to manipulate, mislead, or influence behavior. As AI generated content and deepfakes become increasingly convincing, the line between the two is becoming harder to detect. This creates new risks for governments, businesses, and national security.

Emerging Market Quadrant for Deepfake Detection — Startup Vendors

Gartner frames deepfake detection as an emerging but fast maturing cybersecurity discipline. It is the ability to detect, analyze, and forensically examine fabricated or AI manipulated images, audio, video, and live interactions meant to deceive or disinform. The report urges C-level executives to treat deepfakes as a major cybersecurity risk and evaluates startups on their ability to deliver continuous, cryptographic, forensic grade detection. It splits the vendor landscape into four groups plotted on “Potential to Execute” and “Potential for Market Disruption”: Market Shapers (Reality Defender, GetReal Security, and identifAI), Pace Setters (Resemble AI and Clarity), Pioneers (Netarx, DeepTrust, and Deep Media), and Specialists (pi-labs, Truepic, DuckDuckGoose AI, DeepXL, and Truly).

On strategy, the report identifies three distinct commercialization approaches vendors are pursuing. The first is an ecosystem embedded strategy that weaves multimodal forensic detection directly into clients’ infrastructure via free API access (though this risks commoditization as larger cybersecurity suites absorb baseline detection). The second is a hardware anchored strategy that anchors cryptographic provenance at the physical sensor level for verifiable, tamper proof truth. The final is domain specific interception approach, building highly targeted defenses for specific channels like remote hiring pipelines or live synthetic audio at the endpoint. Looking forward, Gartner expects disinformation security standards (akin to MITRE ATT&CK, via frameworks like DISARM) to emerge, and predicts organizations will build dedicated “digital trust and authenticity platforms” (DTAPs) run by new “TrustOps” teams.

Reality Defender, a New North Ventures portfolio company, is named a Market Shaper, Gartner’s top-tier category for vendors combining the strongest disruptive technology with the strongest go-to-market execution. Market Shapers are distinguished by multimodal ensemble models that scrutinize audio, video, image, and text simultaneously, paired with forensic signals intelligence and threat propagation graphs, embedded into continuous operational workflows. A case example in this quadrant, closely mirroring the executive-impersonation fraud scenario Reality Defender itself has publicized, describes a multinational enterprise targeted by real-time video/audio deepfakes on a live executive videoconference to authorize a fraudulent transfer, resolved by natively embedding multimodal detection into the conferencing platform to catch pixel and audio inconsistencies and block the transaction before execution. Gartner does flag risks even for top-quadrant vendors, though: buyers should watch for platform commoditization, ensure vendors offer genuine proprietary forensic depth rather than easily bypassed pattern recognition, and avoid over-relying on a horizontal, generalized platform if their organization has specialized vulnerabilities better served by domain specific vendors elsewhere in the quadrant.

Aging Pipes, Cyber Threats, and Disinformation: The Battle for America’s Water Security

America’s aging, underfunded water infrastructure, over 2 million miles of pipes, 9-12+ million lead service lines, and widespread PFAS contamination, has become a national security vulnerability, not just an environmental one. Using Jackson, Mississippi’s recurring water crises as a case study, it shows how deferred maintenance, funding gaps, and environmental justice disputes create the kind of divisive, emotionally charged conditions adversaries like China, Russia, and Iran can exploit. The EPA has also documented rising cyberattacks on water systems from actors including Iran’s IRGC, pro Russia hacktivists, and Chinese state sponsored groups like Volt Typhoon, with over 70% of inspected systems found in violation of basic cyber hygiene standards.

The disinformation angle is central: adversaries don’t need sophisticated hacking to cause damage, they can simply amplify real grievances or fabricate false ones about negligence or contamination to erode trust and trigger panic. The article cites a January 2024 incident where a false social media claim that Jackson’s treatment plant had shut down triggered a run on water that overwhelmed the system’s pressure, forcing a boil water notice for 12,000+ residents. That was a single fake post causing real physical strain with no hacking involved. Since water systems are widespread, locally run (93% serve fewer than 10,000 people), and already burdened by real problems tied to race and poverty, they offer a low cost way for adversaries to sow distrust and destabilize communities. The author’s recommendations tie funding, cybersecurity, and information security together, calling for federal agencies to work with state, local, and utility partners on counter disinformation strategies alongside infrastructure investment.

US and South Korea launch first wartime fake news drill

The US and South Korea held their first ever joint tabletop exercise focused specifically on countering foreign disinformation in a wartime scenario. Hosted at the South Korean Joint Chiefs of Staff headquarters in Yongsan, Seoul, the drill brought together the JCS and U.S. Forces Korea, along with personnel from the United Nations Command and the South Korea U.S. Combined Forces Command. Notably, it was the first time military commands partnered with a broad range of civilian government sectors for a unified “whole-of-government” response, with South Korea’s defense, foreign, and culture ministries all helping shape strategies against “cognitive warfare.”

The training centered on three main areas: countering foreign disinformation by identifying and dismantling hostile propaganda in real time; synchronizing allied defenses across cyber, space, and electromagnetic domains; and aligning strategic communications to keep public messaging consistent between allies. Military leaders framed control of the information environment as a top strategic priority for the alliance going forward, and officials are now looking at incorporating these communication and defense strategies into larger live action joint military exercises planned later this year.

More links to explore:

Hubble Pushes Satellite Network For Long-Range Bluetooth

New North Ventures invested in Hubble Network in 2023. Recently Co-Founder and CEO Alex Haro joined Bloomberg Businessweek to discuss the future of Bluetooth and the space economy.

Alex Haro explained how the company has overcome one of the biggest limitations of Bluetooth, its traditionally short range, to enable direct communication with satellites in orbit. Rather than transmitting large amounts of data like headphones or speakers, Hubble optimized Bluetooth for small, high value data packets, such as asset location and status updates, and paired it with proprietary, highly sensitive space based antennas capable of detecting these weak signals from Earth. The result is a network that allows existing Bluetooth devices to connect to satellites without any hardware modifications, dramatically expanding connectivity for Internet of Things (IoT) applications. Haro highlighted commercial use cases across fleet management, construction, mining, oil and gas, and defense, noting that customers like Samsara can enable satellite connectivity through a simple software update. As Hubble continues building out its satellite constellation following launches with SpaceX and other providers, the dual-use company has raised more than $100 million in venture funding and is currently raising additional capital to expand its global network.

Thanks for reading Securing Our Future! Subscribe for free to receive new posts and support our work.

Subscribe now

Can You Grow an Antenna?

Wed, 08 Jul 2026 10:02:46 GMT

DARPA's biological aperture question and the emerging stack for living, programmable RF infrastructure

Aperture is Density

Radio frequency systems are usually described as radios, antennas, radars, and arrays, but the underlying economic variable is aperture. A larger effective aperture can capture more electromagnetic energy, improve directivity, sharpen spatial discrimination, and provide greater control over where a signal is received or projected. Today, that aperture is normally manufactured from metal, dielectric materials, semiconductors, cabling, and precision electronics. The cost and complexity of assembling those components help explain why high performance RF systems remain expensive, power intensive, and difficult to deploy at scale.

DARPA is now asking a question that recasts the problem. On its Ideas Under Incubation page, the agency asks, “Can we create functional bio-apertures with tunable properties?” (Defense Advanced Research Projects Agency [DARPA], n.d., “Biological apertures” section, para. 1). The same prompt asks how biochemical and microbial interventions might enhance metal uptake and sequestration in selected plant species, then connects that biological process to microsystem design. DARPA explicitly labels the page as exploratory rather than a formal solicitation, which makes the idea less a program announcement than an invitation to assemble a new technical field (DARPA, n.d.).

Taken literally, a biological aperture sounds like science fiction. A more disciplined interpretation is a hybrid RF structure in which biology supplies area, growth, morphology, material transport, and some capacity for repair, while engineered materials and conventional electronics provide conductivity, switching, sensing, calibration, and control. No cited experiment demonstrates a field-ready biological radar or phased array. The significance is that several enabling components already exist independently, and their convergence creates a credible research and venture frontier.

The Aperture Is Moving Into the Environment

The first enabling trend is the relocation of RF function from a discrete radio into the surrounding environment. MIT researchers Venkat Arun and Hari Balakrishnan developed RFocus, a six-square-meter surface containing 3,720 inexpensive elements that either transmit or reflect an incident signal. Their central move was conceptual as much as technical: RFocus “moves beamforming functions from the radio endpoints to the environment” (Arun & Balakrishnan, 2019, p. 1). A software controller configured the surface using receiver signal strength measurements, producing a 10.5-fold improvement in median signal strength and a 2.1-fold improvement in median channel capacity in the prototype (Arun & Balakrishnan, 2019).

RFocus was not biological, but it established an important architecture. A small device does not need to carry the entire aperture if the environment can supply area and a controller can organize the resulting electromagnetic behavior. This shifts the commercial question from building a better radio alone to building surfaces, materials, and control systems that make the environment part of the radio. Biological apertures extend that logic by asking whether a portion of the physical surface can be grown, modified, and maintained through living processes.

Plants Already Receive RF Energy

The second enabling trend is direct evidence that living plants can participate in RF systems. Researchers at the Istituto Italiano di Tecnologia modified leaves with a thin elastomeric coating and combined wind-driven contact electrification with radio-frequency energy harvesting. In the paper’s abstract, Meder et al. (2022) report that “the same plant is used as an unmatched Marconi-antenna.” The plant functioned as an ionic receiving antenna for multiband RF energy, and the combined wind and RF harvesting system accumulated enough energy to operate a commercial temperature and humidity sensor that transmitted environmental data wirelessly (Meder et al., 2022).

This experiment should not be confused with a high-gain antenna array. It demonstrated low-power energy conversion, not coherent radar, precision beamforming, or high-rate communications. Its importance lies elsewhere. A common living plant acted as a functional electrical structure across multiple RF bands, integrated with conventional circuitry, and continued to provide biological attributes such as growth and self-repair. That is a foundational proof point for any thesis in which living matter contributes to the physical layer of a sensing or communications system.

Living Systems Can Sense and Report

A third line of research shows that plants can do more than passively interact with electromagnetic energy. Wong et al. embedded carbon-nanotube nanosensors into spinach leaves so that the plants could detect nitroaromatic compounds transported from groundwater into leaf tissue. The researchers described the engineered plants as “infrared communication platforms that can send information to a smartphone” (Wong et al., 2017, Abstract, para. 1). In effect, the plant served as a self-powered sampler, biological transport system, sensor host, and communications interface (Wong et al., 2017).

The relevance to biological apertures is architectural rather than frequency specific. The nanobionic spinach did not transmit conventional RF signals, but it demonstrated that a living organism can gather materials from its environment, concentrate them through its vascular system, host engineered nanoscale components, and communicate a resulting measurement to external electronics. DARPA’s interest in enhancing metal uptake and sequestration suggests a related systems logic: use biological transport and accumulation to place electromagnetically useful material inside or along a living structure, then connect that structure to microsystems that make the material operational.

RF Materials No Longer Need to Look Like RF Hardware

A fourth enabling trend comes from additive manufacturing and conductive polymers. Yurduseven et al. built a holographic metasurface antenna for 10 GHz operation using a biodegradable conductive polymer for the conducting regions and polylactic acid for the dielectric substrate. The authors emphasize that “The entire metasurface antenna is 3D printed at once” (Yurduseven et al., 2018, p. 1). Their experiments demonstrated high-fidelity focusing in the antenna’s Fresnel region, while also showing that material conductivity remained a critical determinant of radiation performance (Yurduseven et al., 2018).

The result matters because it separates RF function from the assumption that every aperture must be machined from conventional metal assemblies. Conductive polymers, printed structures, embedded nanoparticles, and biologically accumulated metals offer a wider material design space. Biology will not eliminate the conductivity requirement, and low-conductivity materials impose real efficiency penalties. It may, however, supply inexpensive, conformal, and spatially distributed scaffolding onto which electromagnetically functional materials can be deposited, transported, or organized.

Growth Can Become a Fabrication Process

The fifth enabling trend is partial control over biological growth. The flora robotica project combined natural plants, robotic scaffolds, sensors, light, and hormones to explore living architectural systems with continuous growth and self-repair. Hamann et al. (2017) explain that “User-defined design objectives help to steer the directional growth of the plants” (p. 1). The project used phototropic and shade-avoidance responses to guide climbing plants through mechanical structures, demonstrating that growth can be influenced toward desired geometries even when the process remains slower and less deterministic than industrial fabrication (Hamann et al., 2017).

That distinction is essential. A future biological aperture would not emerge from the ground with the precision of a lithographically produced array. The more realistic model is guided growth around a scaffold, followed by measurement, functionalization, and computational calibration. The biological structure would provide area and evolving geometry. Conventional engineering would impose enough order to make that imperfect structure useful.

What the First Biological Aperture Would Probably Look Like

A plausible first-generation bio-aperture, inferred from these research lines, would be hybrid at every layer. A plant, fungal network, or engineered living material could provide a large and conformal substrate. Biochemical interventions could increase the uptake or localization of selected metals. Conductive polymers, coatings, or nanoparticles could create more reliable current paths. Small conventional nodes could provide impedance matching, switching, sensing, telemetry, and power management. A software controller could maintain a digital model of the structure and compensate for changes in geometry, hydration, temperature, growth, or material distribution (DARPA, n.d.; Hamann et al., 2017; Meder et al., 2022; Yurduseven et al., 2018).

The first useful products would probably be passive or low power. Candidate functions include environmental spectrum monitoring, RF energy harvesting, frequency-selective surfaces, low-power reflectors, conformal receiving structures, or distributed sensing around farms, forests, buildings, borders, and critical infrastructure. Coherent transmission, high-power operation, and precision radar would require much tighter control over conductivity, phase, element spacing, feed networks, heat, and calibration. The near-term opportunity is therefore not a tree that replaces an exquisite radar. It is a new class of living or bio-derived surfaces that contribute useful area and electromagnetic function to a conventional system.

Where the Venture Stack Could Form

The biological layer offers one potential company-formation point. Tools that control metal uptake, transport, localization, and sequestration could become enabling platforms for more than RF, including environmental remediation, biomining, sensors, and advanced materials. The investable moat would likely reside in repeatable control over where functional material accumulates, how concentrations are measured, and how the process performs across species and field conditions. DARPA’s framing suggests that biology is not merely a substrate; it may be part of the manufacturing process itself (DARPA, n.d.).

The materials and interface layer offers a second opportunity. Tissue compatible conductive polymers, nanoparticles, coatings, electrodes, and low-power microsystems will be required to bridge wet, variable biological matter with dry, precision electronics. The winning architecture may be modular: biology supplies area and material transport, while standardized interface nodes supply switching, matching, calibration, and secure communications. This layer could mature before a complete biological aperture because each component also serves adjacent markets in wearables, agriculture, environmental sensing, soft robotics, and distributed infrastructure.

The computational layer may be the most leveraged. A biological aperture would be a moving, wet, heterogeneous RF object whose geometry and material properties change over time. Each deployed system could be slightly different. That complexity increases the value of electromagnetic simulation, inverse design, digital twins, synthetic RF data, adaptive calibration, and control software. RFocus showed that software can organize thousands of simple environmental elements without requiring each element to be an exquisite RF component (Arun & Balakrishnan, 2019). A biological system would push that principle further: imperfections would not be removed entirely through manufacturing, but measured and compensated for computationally.

The Investment Takeaway

DARPA’s biological-aperture question matters because it is an integration roadmap disguised as a speculative prompt. Researchers have already shown that beamforming can move into the environment, that plants can receive multiband RF energy, that living plants can host engineered sensors and communicate externally, that conductive polymers can form working microwave apertures, and that biological growth can be guided toward designed structures. None of these results proves that a useful biological aperture exists today. Together, they show that the category is no longer pre-science.

The first venture-scale winner may not market a biological antenna. It may sell the metal-uptake control system, the conductive biointerface, the embedded microsystem, the electromagnetic digital twin, or the calibration software that makes irregular living materials reliable. The enduring investment question is therefore not whether a tree can replace a radar. It is whether biology can become a programmable manufacturing and deployment layer for RF infrastructure, and which company will own the control point that makes that transition repeatable.

References

Arun, V., & Balakrishnan, H. (2019). RFocus: Practical beamforming for small devices. arXiv. https://doi.org/10.48550/arXiv.1905.05130

Defense Advanced Research Projects Agency. (n.d.). Ideas under incubation. Retrieved June 21, 2026, from https://www.darpa.mil/research/ideas

Hamann, H., Divband Soorati, M., Heinrich, M. K., Hofstadler, D. N., Kuksin, I., Veenstra, F., Wahby, M., Nielsen, S. A., Risi, S., Skrzypczak, T., Zahadat, P., Wojtaszek, P., Støy, K., Schmickl, T., Kernbach, S., & Ayres, P. (2017). Flora robotica: An architectural system combining living natural plants and distributed robots. arXiv. https://doi.org/10.48550/arXiv.1709.04291

Meder, F., Mondini, A., Visentin, F., Zini, G., Crepaldi, M., & Mazzolai, B. (2022). Multisource energy conversion in plants with soft epicuticular coatings. Energy & Environmental Science, 15, 2545-2556. https://doi.org/10.1039/D2EE00405D

Wong, M. H., Giraldo, J. P., Kwak, S.-Y., Koman, V. B., Sinclair, R., Lew, T. T. S., Bisker, G., Liu, P., & Strano, M. S. (2017). Nitroaromatic detection and infrared communication from wild-type plants using plant nanobionics. Nature Materials, 16, 264-272. https://doi.org/10.1038/nmat4771

Yurduseven, O., Ye, S., Fromenteze, T., Marks, D. L., Wiley, B. J., & Smith, D. R. (2018). 3D conductive polymer printed metasurface antenna for Fresnel focusing. arXiv. https://doi.org/10.48550/arXiv.1806.00394

More links to explore:

Exclusive: Auriga Launches Hypersonic Weather Testing Chamber

New North Ventures portfolio company Auriga Space has launched a specialized hypersonic weather testing chamber designed to accelerate the development and validation of materials and coatings used in hypersonic weapons systems. The chamber replicates extreme environmental conditions, including temperature, humidity, pressure, and atmospheric composition, that hypersonic systems encounter during flight.

As the government pushes to field hypersonic capabilities at speed, testing infrastructure has become a critical bottleneck, particularly for unsexy but mission critical components like thermal protection systems, sensor windows, and structural coatings. Auriga's facility addresses that gap by providing rapid, repeatable testing that can validate designs earlier in the development cycle, reducing risk and compressing timelines for programs under intense pressure to deliver operational capability.

Thanks for reading Securing Our Future! Subscribe for free to receive new posts and support our work.

Subscribe now

Orbital Infrastructure: Computing Moves to Space

Wed, 01 Jul 2026 10:00:46 GMT

Orbital files plans for 100,000 orbital data centers

Five-month-old startup Orbital has asked the Federal Communications Commission for permission to deploy up to 100,000 data center satellites, aiming to bring 10 gigawatts of computing power from space to meet rising artificial intelligence demand. The filing represents one of the most ambitious orbital infrastructure proposals to date and signals a fundamental shift in how the defense and commercial sectors are thinking about compute infrastructure. Rather than viewing satellites as communications or sensing platforms, Orbital is positioning space as the next frontier for raw computational capacity, particularly for power intensive AI workloads that are increasingly constrained by terrestrial energy, cooling, and regulatory limitations.

The scale is striking. At 100,000 satellites and 10 gigawatts of power, Orbital is proposing an orbital compute constellation that rivals the entire terrestrial hyperscale data center footprint of a major cloud provider. The company is betting that space offers unique advantages for AI training and inference: near unlimited solar power, natural thermal management through radiative cooling, reduced latency for globally distributed workloads, and insulation from terrestrial infrastructure vulnerabilities. This is not incremental optimization. Orbital compute is being designed explicitly to serve AI workloads, which are the single largest driver of infrastructure demand in both commercial and defense contexts today.

The broader implication for defense investing is that the boundary between space infrastructure, AI, and national security is collapsing. Orbital compute constellations will be attractive to defense and intelligence users for workloads that require airgapped processing, sovereign compute isolation, or persistent overhead presence. They also introduce new categories of vulnerability and resilience, from space based cyber defense to on-orbit fault tolerance. For early stage investors, the signal is clear: the next generation of defense infrastructure will not be built on the ground, it will be built in orbit. The companies enabling that shift, whether through launch, power systems, thermal management, networking, or the compute platforms themselves, represent the infrastructure layer beneath every future defense AI capability.

Buckle Up: The Bad Guys Now Have A Model As Powerful As Mythos

A significant shift has occurred in AI powered cybersecurity: China has released an open weight AI model with offensive cyber capabilities comparable to Anthropic’s restricted Mythos model. Unlike Mythos, which remains tightly controlled because of its ability to autonomously discover software vulnerabilities and generate exploits, the Chinese model is publicly available. This means nation states, cybercriminals, and other malicious actors can freely access advanced offensive cyber capabilities. This could dramatically lowers the barrier to conducting sophisticated cyberattacks and marks a new phase in the AI arms race, where powerful cyber tools are no longer confined to a handful of leading AI labs.

The article also highlights the growing geopolitical implications of frontier AI. While U.S. companies have emphasized strict safeguards around their most capable models, China’s decision to openly release a similarly powerful system underscores the tension between AI safety and strategic competition. Restricting access to advanced models in democratic countries may no longer prevent misuse if comparable capabilities are readily available elsewhere. As a result, governments, critical infrastructure operators, and enterprises will need to accelerate investments in AI enabled cyber defense, assuming that attackers now have access to tools capable of discovering and exploiting vulnerabilities at unprecedented speed and scale.

NASA races to save Swift telescope from falling back to Earth with daring rescue mission

NASA is launching a first-of-its-kind $30 million robotic satellite servicing mission to rescue its aging Neil Gehrels Swift Observatory, which has been in orbit since 2004. Increased solar activity has expanded Earth’s upper atmosphere, creating more drag and causing Swift’s orbit to decay much faster than expected. To prevent the telescope from reentering the atmosphere, startup Katalyst Space Technologies will launch its LINK servicing spacecraft aboard a Northrop Grumman Pegasus XL rocket. LINK will autonomously rendezvous with Swift, capture it using robotic arms, and gradually raise its orbit from roughly 224 miles to 373 miles, extending the observatory’s operational life by several years.

Beyond saving a scientifically valuable telescope, the mission represents a major milestone for the emerging on-orbit servicing industry. Rather than replacing expensive space assets, operators may soon be able to repair, refuel, or reposition satellites already in orbit. If successful, the Swift rescue would move orbital servicing beyond one time demonstrations into a practical commercial capability, paving the way for future life extension missions for satellites.

More links to explore:

Psionic Selected by NASA Under the 2025 Announcement of Collaboration Opportunity (ACO)

Psionic has been selected by NASA as one of 37 U.S. companies participating in the agency’s 2025 Announcement of Collaboration Opportunity (ACO), a program that gives innovative commercial companies access to NASA’s facilities, hardware, software, and technical expertise to accelerate technologies for future lunar and Mars missions. The collaboration underscores NASA’s continued investment in next generation commercial space capabilities.

Through the partnership, Psionic will work with NASA’s Armstrong Flight Research Center to flight test its Navigation Doppler Lidar (NDL) technology, which enables precise navigation in GPS denied environments by measuring velocity and range. Successful validation could help advance autonomous navigation systems for future lunar landers and human exploration missions to the Moon and Mars.

Thanks for reading Securing Our Future! Subscribe for free to receive new posts and support our work.

Subscribe now

Critical Minerals In Focus

Wed, 24 Jun 2026 10:01:05 GMT

Critical minerals have moved from the margins of industrial policy to the center of national strategy, and for good reason. From the lithium powering electric vehicle batteries to the rare earth elements embedded in fighter jet components and semiconductor manufacturing, these materials underpin virtually every sector that defines modern technological and military power. As artificial intelligence, clean energy, and advanced defense systems continue to scale, the demand for minerals like manganese, nickel, cobalt, copper, and rare earths is accelerating and the nations that control their supply are gaining enormous geopolitical leverage. For the United States, the stakes are stark: over reliance on foreign sources, particularly China, which dominates global rare earth processing, has exposed critical vulnerabilities in domestic supply chains that touch everything from consumer electronics to missile guidance systems. Securing a stable, domestic supply of critical minerals is now a matter of national security, technological sovereignty, and the country’s ability to compete and innovate.

Trump Administration Advances Offshore Critical Minerals Development Off Virginia

The U.S. Bureau of Ocean Energy Management (BOEM) is advancing its evaluation of offshore critical mineral development off the coast of Virginia by issuing a Request for Information and Interest (RFI) to gauge industry appetite and gather public input on potential leasing areas. The notice opens a 30 day public comment period and is an early step in BOEM’s planning process.

The initiative is part of the Trump administration’s broader effort to expand domestic production of critical minerals used in manufacturing, defense systems, and emerging technologies, with the goal of reducing dependence on foreign supply chains. BOEM Acting Director Matt Giacona framed it as a way to strengthen America’s strategic position through secure domestic supply chains.

The effort follows BOEM’s December 2025 announcement that it was considering offshore mineral leasing off Virginia after receiving an unsolicited lease sale request from Florida based Odyssey Marine Exploration, which has expressed interest in developing heavy mineral sands and phosphorite deposits on the U.S. Outer Continental Shelf. That announcement was only the third offshore minerals leasing action of its kind in more than three decades, and came after President Trump’s executive order directing federal agencies to accelerate domestic development of strategic mineral resources. Public comments on the RFI will be accepted through July 23 via Regulations.gov.

DOD commits $1.2B in conditional loans to Phoenix Tailings, Energy Fuels

Phoenix Tailings, a Massachusetts based rare earth processing company with an operational facility in Exeter, New Hampshire, received a conditional $500 million loan commitment from the Department of Defense’s Office of Strategic Capital (OSC). The financing, combined with additional private capital, is expected to support an approximately $1 billion effort to expand domestic rare earth processing capacity and build the company’s planned “Freedom Facility.” This will be a large scale separation and metallization plant designed to strengthen the U.S. critical minerals supply chain.

The investment highlights growing federal urgency around re-shoring critical mineral and rare earth processing capabilities that are currently dominated by China. Phoenix Tailings has developed technology to process rare earth materials from mined, recycled, and secondary feedstocks. The expansion is intended to support U.S. defense, energy, advanced manufacturing, and permanent magnet supply chains while reducing dependence on foreign processing infrastructure.

China Targets US Rare Earth Firms in Response to Pentagon

China escalated its rare earth supply chain response to ongoing U.S.-China trade and national security tensions by adding two leading American rare earth companies, MP Materials and USA Rare Earth, to its export control list. The move restricts Chinese companies from exporting dual-use goods to the affected firms and underscores Beijing’s willingness to leverage its dominance in rare earth processing as a geopolitical tool. While the immediate operational impact may be limited, the action highlights the strategic importance of building domestic mine-to-magnet supply chains and reducing U.S. dependence on Chinese rare earth materials and processing capacity.

The inclusion of MP Materials, which operates the only active rare earth mine in the United States, and USA Rare Earth, which is building an integrated domestic rare earth and magnet manufacturing platform, reinforces the growing significance of rare earths as a national security priority.

More links to explore:

HawkEye 360 Looks Overseas For Its Next Growth Leg

HawkEye 360 is increasingly looking beyond the U.S. government market for its next phase of growth, expanding sales to allied nations that are seeking sovereign space based intelligence capabilities. The company’s radio frequency (RF) intelligence platform enables customers to detect and geolocate signals from ships, aircraft, GPS jammers, and other emitters worldwide, providing critical situational awareness without relying on traditional intelligence infrastructure. International demand has accelerated amid rising geopolitical tensions and growing defense spending across Europe, the Indo-Pacific, and other allied regions.

The strategy appears to be gaining traction, with HawkEye 360 reporting more than $100 million in new international contract awards this year, including a significant electronic warfare program with a European Ministry of Defense. As governments increasingly prioritize space based ISR, electronic warfare, and spectrum intelligence capabilities, HawkEye 360 is positioning itself as a key provider of commercial signals intelligence to both U.S. and allied defense customers worldwide.

Thanks for reading Securing Our Future! Subscribe for free to receive new posts and support our work.

Subscribe now

The Moat and Lessons Learned From Industry

Wed, 17 Jun 2026 10:00:37 GMT

Defensibility in Defense Tech: What Actually Counts as a Moat

The question to ask nearly every founder, and if you’re a founder, ask yourself, is this: “if this works, what stops the next ten teams, and the primes, from doing the exact same thing in eighteen months?”

If the honest answer is “nothing but execution,” you’re looking at a feature, not a company. If the answer is “years of qualification, a proprietary process, a supply chokepoint, or physics most teams simply can’t replicate,” then we can start talking about a moat.

So what is a moat? Warren Buffett’s framing still holds: the business is a castle, and the moat is what keeps attackers out. In venture, you’re not underwriting whether something is good today, you’re underwriting whether returns survive once the market realizes it’s a good market. Without a moat, margins collapse as competition floods in. A moat is what prevents capital and talent from simply taking your customers.

The classic sources still apply such as IP, switching costs, network effects, scale advantages, regulatory barriers. However, defense bends them and adds a few that the commercial world rarely has:

Accreditation – One of the most underrated barriers in the sector. Clearances, FedRAMP High, DoD IL5/IL6, ATOs, weapons qualification. These take years and millions in “paperwork” that is actually infrastructure. You cannot fake a cleared workforce or accredited facility; it is a decade long build. Palantir is one of only a handful of companies, alongside Microsoft and AWS, cleared at DoD IL6. That’s not a feature; it’s a locked gate.

Programs of Record – Defense procurement is sticky in a way commercial markets aren’t. Win a program of record and you become the baseline for 10–20 years. Switching costs aren’t churn, they’re full system re-architecture.

Physics and production – If you can deliver something cheaper than anyone else, competition doesn’t erode you, it breaks everyone else. This is the SpaceX story, and increasingly true in maritime and munitions.

Data – Whoever becomes the system of record owns the workflow. Once you define the data model, ripping it out means rebuilding the entire operating picture.

Capital – Cost-plus primes have little incentive to move fast. Startups that fund R&D on their own balance sheet can out-iterate incumbents, until that speed is either converted into a structural moat or it disappears.

Now let’s look at a few winners, because each illustrates a different moat and most stack multiple.

SpaceX

The lazy take is that SpaceX’s moat is reusable rockets. It isn’t anymore. Blue Origin landed an orbital class booster in 2025, and others are close, proving the physics is replicable. The real moat is compounded operational learning that only cadence can create.

SpaceX flew ~170 times in 2025 across nearly 500 total missions, driving cost per kilogram down more than 95%. Every flight improves refurbishment, reliability, and mission design in ways competitors can’t replicate without the same flight volume. The flywheel reinforces itself: cheap launch enables Starlink, Starlink fills the manifest, and a full manifest funds more cadence.

On top of that sits Starlink: ~10 million subscribers and $10B+ in 2025 revenue, with improving margins. That creates a network and capital moat layered on launch dominance. Then comes government entrenchment: NSSL backlog, NASA contracts, and Starshield positioning it as default infrastructure for national security space access.

As SpaceX approaches a ~$1.77T valuation in private markets, the signal is clear: these moats compound.

Lesson: durable advantage isn’t a feature, it’s a loop that gets stronger every time it runs.

Palantir

Palantir is often misread as an AI company. It isn’t. It uses other people’s models such as OpenAI, Anthropic, and others.

The real moat is the data integration layer: Foundry and Gotham. It is a semantic ontology that maps how an organization’s fragmented data actually fits together. Building it takes 6–18 months. Once it exists, switching AI vendors is trivial but rebuilding the ontology elsewhere is a multi-year reintegration effort most organizations won’t attempt.

That layer becomes embedded in daily operations. Removing it would break the system.

Wrapped around it is accreditation: FedRAMP High, IL5/IL6, and a cleared workforce. Palantir turned that burden into a product with FedStart, letting others operate inside its accredited environment instead of rebuilding their own ATO path. That converts compliance into a platform advantage.

The moat is the integration layer and it outlives whatever model sits on top of it.

Lesson: be the system of record, not the application. Apps get replaced. Data models do not.

Anduril

Anduril is often described as a modern prime, but it is structurally anti-prime. Traditional primes rely on cost-plus contracts that reward neither speed nor efficiency. Anduril flipped the model: hardware is the delivery mechanism for a software platform.

Lattice is the core: it fuses sensors, drones, towers, and effectors into a unified operating system. The result is a vertically integrated system with a horizontal software engine, one backbone across many hardware classes.

The second moat is speed-as-capital. By funding R&D privately, Anduril reaches deployment years ahead of traditional procurement cycles. That’s how it secured a $20B Army enterprise deal consolidating 120+ procurement pathways around Lattice.

The third moat is emerging: Lattice as a standard. As it becomes the integration layer, competitors either plug in or get excluded from systems built around it.

Lesson: if you sell hardware, find the recurring software layer or become the standard others must integrate with.

How to build real moats

Start the boring barriers early. Clearances, ATOs, accreditation, qualification. Build them before you need them. They look like cost today and a moat tomorrow.

Use capital and speed deliberately. Outspend incumbents on R&D and convert that speed into structural advantage before the capital edge fades. Speed is a moat with an expiration date.

Don’t confuse contracts with moats. A contract is an outcome. A moat is what ensures the next one.

The defense boom is real, and it’s producing winners faster than ever. But the companies pulling away all share the same pattern: compounding cadence, integration layers others can’t unwind, and standards everyone else must adopt.

Everyone has a contract. The question is: who has a castle.

Anthropic Disabled Fable 5 And Mythos 5 After A U.S. Export-Control Order. Here What Happened

Anthropic, one of the leading frontier AI labs, has disabled access to its most advanced models, Fable 5 and Mythos 5, following a new U.S. export control order that restricts foreign access to cutting edge artificial intelligence systems. The company is now engaged in ongoing negotiations with the Trump administration to restore access, but so far no breakthrough has been reached. The move represents one of the most significant exercises of AI export restrictions to date, affecting not only international users but also raising questions about how compute thresholds, model capabilities, and national security considerations will intersect as the technology continues to advance.

The timing and scope of the restrictions are notable. Anthropic’s Fable 5 and Mythos 5 are widely considered to be among the most capable large language models available, used across enterprise, research, and government applications. By targeting these specific systems, the U.S. government is signaling that it views advanced AI as dual-use infrastructure requiring the same kind of strategic export discipline traditionally applied to semiconductor manufacturing equipment, cryptography, or missile technology. This is no longer theoretical policy. It is operational doctrine, and it marks a shift from voluntary corporate governance to direct state intervention in the AI supply chain.

More links to explore:

Hubble Network, InPlay Partner to Bring Sub-$1 Global Tracking

Hubble Network and InPlay have partnered to make global, item level asset tracking as inexpensive as passive RFID by combining InPlay’s sub $1 NanoBeacon chip with Hubble’s satellite powered Bluetooth network. The solution allows smart labels, wearables, and IoT sensors to transmit location and condition data worldwide without requiring companies to deploy readers, scanners, cellular modems, or other tracking infrastructure. The partnership is aimed at supply chain visibility, cold chain monitoring, returnable transport assets, and other use cases where traditional RFID lacks range and cellular tracking is too expensive, potentially enabling low cost tracking of billions of everyday items.

Thanks for reading Securing Our Future! Subscribe for free to receive new posts and support our work.

Subscribe now

Maritime Autonomy: Venture Scale Investment

Wed, 10 Jun 2026 10:03:58 GMT

The Venture Capital Case for Maritime Autonomy

When we look at the maritime industry, we like to break it down into two main categories: unmanned undersea and unmanned surface maritime technology, where both categories include hardware and software solutions. What has changed over the last several years is that maritime autonomy has evolved from a niche defense market into a venture scale investment category. Growing geopolitical competition, particularly in the Indo-Pacific, the Red Sea, and other contested waterways, has transformed both the ocean surface and undersea environment into critical zones of military competition. As a result, governments are allocating significantly more resources and capital toward autonomous maritime capabilities that can provide persistent surveillance, logistics, mine countermeasures, anti-submarine warfare, and distributed lethality at a fraction of the cost of traditional crewed platforms. The U.S. Navy alone continues to expand investments in unmanned maritime systems and broader fleet modernization efforts.

The buyer universe has also expanded considerably. In addition to traditional defense customers such as the U.S. Navy, SOCOM, and allied maritime forces, strategic acquirers now include large defense primes, shipbuilders, autonomy software providers, and dual-use technology companies seeking to operate across land, air, sea, and undersea domains. This is creating a much more active merger and acquisition environment as companies race to build all-domain autonomous capabilities rather than develop every component internally. Recent examples include Havoc AI’s expansion of its maritime autonomy platform to land, sea and air, reflecting a broader industry trend toward acquiring or integrating technologies that enable cross-domain operations.

Capital formation is accelerating alongside acquisition activity. Havoc AI recently raised $85 million, bringing total funding to nearly $100 million in less than two years, while Blue Water Autonomy raised a $50 million Series A after an earlier $14 million seed round. Booz Allen recently invested in maritime company Ulysses as part of their recent Series A round. Other strategic investors such as Lockheed Martin, Hanwha, and In-Q-Tel are increasingly investing directly into maritime autonomy companies, signaling both customer demand and future acquisition interest. As autonomous maritime systems prove operationally effective and become programs of record, we expect acquisition timelines to compress, strategic premiums to increase, and consolidation to accelerate. Similar to what occurred in the drone and defense software markets over the past decade, maritime autonomy is rapidly becoming a core layer of future military capability and one of the most attractive emerging categories for venture backed defense technology.

We have recently seen several interesting companies and wanted to highlight some companies that are building in the maritime space below. If you are a pre-seed or seed company building in this domain, reach out! We would like to learn more.

Op-Ed: Maritime Cybersecurity is a Financial Investment

The article frames maritime cybersecurity as a core financial investment rather than a discretionary IT cost. Modern shipping and port operations are deeply digitized, with vessels relying on interconnected systems for navigation, engine control, cargo handling, communications, and port coordination. This connectivity expands the attack surface, making ships and maritime infrastructure increasingly attractive targets for cybercriminals and state actors. The author emphasizes that even relatively simple intrusions, such as tampering with AIS data or accessing onboard systems, can create real operational disruption and expose sensitive commercial and security information.

The central argument is that the economics of cyber risk in maritime are now impossible to ignore. A single cyber incident can lead to costly vessel downtime, delayed cargo, supply chain disruption, and reputational damage that cascades across global logistics networks. Yet many operators still underinvest in cybersecurity relative to the scale of exposure. The piece contends that treating cybersecurity as a line item operating expense underestimates its return profile: modest, consistent investment in cyber defenses can prevent disproportionately large losses, reduce operational risk, and strengthen resilience across increasingly software-driven maritime systems.

More links to explore:

ZeroFox and Reality Defender Partner to Close the Synthetic Media Gap in External Threat Protection

ZeroFox and Reality Defender have formed a partnership to address the growing risk of AI generated “synthetic media” being used in cyberattacks. The integration embeds Reality Defender’s deepfake detection capabilities directly into the ZeroFox external threat protection platform, allowing security teams to identify whether images, audio, or video content used in impersonation and social engineering campaigns has been artificially generated. The goal is to close a gap in traditional external threat workflows, which can detect malicious infrastructure but often cannot verify the authenticity of media itself. By adding real time multimodal detection inside ZeroFox’s existing “Discover, Validate, Disrupt” workflow, analysts can more quickly confirm and respond to impersonation attempts involving executives, brands, or other high value targets. The companies argue this will improve accuracy, reduce time to takedown, and help organizations better defend against increasingly realistic AI driven fraud and misinformation campaigns.

Thanks for reading Securing Our Future! Subscribe for free to receive new posts and support our work.

Subscribe now

Battlefield AI: The Emerging Doctrine Divide

Wed, 03 Jun 2026 10:01:09 GMT

As Pentagon pushes for battlefield AI, some military leaders urge caution

The Pentagon’s aggressive push toward battlefield AI is meeting resistance from within its own ranks, creating a significant doctrinal divide over the role of autonomous decision making in combat operations. While the Department of War continues to accelerate AI integration across targeting, reconnaissance, and command systems, a growing faction of senior military leaders is publicly urging restraint, particularly around lethal autonomous weapons and AI driven strike decisions. The tension reflects a broader uncertainty about the operational, ethical, and strategic implications of delegating life and death choices to algorithmic systems in high stakes environments.

What makes this moment notable is not the presence of skepticism, which has always existed around emerging military technologies, but rather the timing and seniority of those voicing concern. This is active duty and recently retired flag officers questioning whether the current pace of AI deployment is outstripping the military’s ability to validate performance, understand failure modes, or maintain meaningful human control in dynamic combat scenarios. The concerns span technical reliability, adversarial manipulation, escalation risks, and the erosion of human judgment in contexts where context, culture, and proportionality matter as much as precision.

The broader implication for defense technology investors is that the pathway from prototype to production is becoming more contested, even within the customer base. Autonomous systems companies can not assume that capability demonstrations alone will unlock procurement at scale. There is a widening gap between what the technology can do and what military institutions are organizationally, legally, and culturally prepared to adopt. The winners in this market will be companies that build not just performance, but verifiability, explainability, and graduated autonomy into their systems. Those that can help the Defense Department navigate the gap between ambition and doctrine will capture the long term value, while those that push purely on capability may find themselves technically impressive but operationally sidelined.

US, UK, Australia launch joint underwater drone project by 2027

The United States, United Kingdom, and Australia have announced a joint initiative to develop and field underwater autonomous systems by 2027 under the AUKUS security framework. The program aims to accelerate the deployment of unmanned undersea vehicles for intelligence, surveillance, reconnaissance, and potentially offensive missions across contested maritime domains. The collaboration will focus on interoperability, shared development timelines, and coordinated procurement pathways. This signals that underwater autonomy is becoming a trilateral strategic priority rather than a series of parallel national programs.

The move reflects growing recognition that the subsurface domain is becoming as contested and technologically dynamic as airspace. Undersea drones offer persistent monitoring of critical infrastructure, submarine tracking, mine countermeasures, and distributed naval presence without the cost, risk, or manpower requirements of crewed platforms. For dual use investors, the signal is clear: underwater autonomy is transitioning from experimental capability to core defense infrastructure, with allied interoperability as a forcing function for standardization and scale. Companies building modular, software defined underwater platforms with open architectures stand to benefit as AUKUS procurement accelerates over the next 24 months.

America’s Next Arsenal Will Be Built by Startups

A new narrative is solidifying in Washington and across defense policy circles: the next generation of American military capability will not come primarily from traditional primes, but from venture backed startups building software driven, modular, and rapidly iterating platforms. The argument centers on speed, cost, and adaptability. Startups are fielding drones, AI targeting systems, counter UAS platforms, and autonomous vehicles in months rather than years, at unit costs orders of magnitude below legacy programs. Lessons from Ukraine, where commercial and startup built systems have defined battlefield outcomes, are accelerating this shift in procurement philosophy.

What matters here is not just the technology, but the industrial and cultural transformation this represents. Defense startups are forcing the Pentagon to rethink how it buys, tests, and integrates new capabilities. Companies like Anduril, Shield AI, and dozens of others are building not just products but entirely new models of defense production, ones rooted in continuous software updates, modular hardware, and venture scale rather than cost plus contracting. For investors, this is the industrial base thesis playing out in real time. The companies that can combine technical performance with the operational maturity to navigate government bureaucracy, security requirements, and allied interoperability will define the next decade of defense innovation. Those that cannot will remain impressive demos without pathways to production.

More links to explore:

Auriga Space | California firm rethinks missile defense without rocket motors

Auriga Space is developing a novel missile defense concept that replaces traditional rocket powered interceptors with electromagnetic launch systems capable of accelerating projectiles to hypersonic speeds using electricity. The approach aims to address a growing challenge in U.S. air and missile defense: interceptor production is constrained by reliance on solid rocket motors and a fragile domestic supply chain for key propellants. Backed by contracts from the Missile Defense Agency and AFWERX, Auriga argues that reusable electromagnetic launchers could dramatically lower the cost per interception while reducing dependence on scarce rocket components. The company plans to begin hypersonic testing in 2026, with the long term vision of deployable, containerized launch systems that could provide more scalable and affordable missile defense against drones, missiles, and other emerging threats.

HawkEye 360 draws broad Wall Street support following IPO debut

Space based signals intelligence company HawkEye 360 is receiving strong attention from Wall Street following its recent IPO, with several major banks initiating coverage and most assigning Buy, Outperform, or Overweight ratings. Investors are drawn to the company’s unique position as a commercial provider of radio frequency (RF) geospatial intelligence, using a constellation of more than 30 satellites to detect and analyze signals from radar systems, communications networks, maritime vessels, and GPS jammers. Analysts see HawkEye as a beneficiary of rising global defense spending, growing demand for space based intelligence, and the shift of ISR missions from traditional airborne platforms to satellites. While some firms caution that the stock’s valuation already reflects much of its growth potential, the broader consensus is that HawkEye is well positioned to capitalize on an expanding national security space market expected to grow significantly over the coming decade.

Thanks for reading Securing Our Future! Subscribe for free to receive new posts and support our work.

Subscribe now

Boston Tech Week & Where to Build a Defense StartUp

Wed, 27 May 2026 10:01:37 GMT

It’s Boston Tech Week, which means hundreds of founders, investors, operators, and researchers are gathering across the city to talk about what’s next. But beyond the noise, the real opportunity is finding the signal and increasingly that signal is defense tech, dual-use innovation, and the builders creating companies in Boston. Few ecosystems combine the same density of world class labs, research universities, engineering talent, maritime testing grounds, early stage accelerators, and government connected innovation networks. Boston has become a place where companies can go from breakthrough research to deployment-ready technology faster than almost anywhere else.

This Thursday, May 28, New North Ventures Co-Founder & GP Jeremy Hitchcock will be speaking at Boston Tech Week on the panel “Future of Dual-Use: What’s Pacing Human-Out-Of-The-Loop Autonomy?” As more founders realize Boston is one of the best places in the country to start and scale frontier technology companies, conversations around autonomy, defense innovation, and national security are becoming central to the region’s next chapter of growth. We hope to see you around Boston this week and reach out if you are a founder building in defense tech.

Where to Build A Defense Start Up

Most founders default to Washington, DC as the place they need to be to build their tech. They think about proximity to Congress, to lobbyists, to the prime contractors with the long institutional foundations. They’re not wrong that DC matters. They’re wrong about how much it matters and when it matters.

Where you build a defense startup is a capital allocation decision. It shapes who you can hire, how fast you can iterate, who will fund you, and how quickly you can get in front of the operators who will actually use what you build. Getting it wrong doesn’t kill you outright, it just adds drag, at every stage, compounding over time.

Here’s our perceived map of where we think are some hubs, while keeping in mind that in today’s day and age, if your technology is great, you can be anywhere and investors will find you.

Los Angeles, CA:

This is where a defense tech renaissance was, over the past two decades, born and where it remains densest in raw industrial terms. Anduril (Costa Mesa), SpaceX (Hawthorne), Impulse Space (Redondo Beach) among others.

The clear connection here is manufacturing gravitas. At least half of all space vehicles and satellites are currently manufactured in El Segundo. The Los Angeles area has decades of aerospace talent, which includes propulsion engineers, systems integrators, manufacturing workers who have seen generations of defense tech pass by before their eyes. They’ve been building defense tech for Northrop, Raytheon and Boeing for years. The result of this is a talent density that’s nearly impossible to replicate.

For founders building in hardware intensive defense domains, to include autonomous systems, propulsion, weapons systems and spacecraft, there is probably no better place. The ecosystem has also produced a network of ex-SpaceX, ex-Palantir, ex-Anduril engineers who are now founders themselves, creating a secondary talent gravity that accelerates startup formation.

NOVA/DC:

Sure, we mentioned DC earlier, and it might’ve felt like we were tossing shade at it but in all realities, it’s the oldest answer in the define startup playbook. For the right type of company, it’s definitely still the answer, it’s just not the only answer anymore. Companies in the NCR have proximity to the Pentagon, the intelligence community, and the defense contracting ecosystem, making it the rational choice when your go-to-market depends on deep government relationships, classified environments, and long acquisition cycles.

The structural advantage is customer density. In DC/NOVA, the program offices that typically decide whether to buy your technology, the contracting officers who execute the deals, and the end users who will actually operate your system can all be reached within a 30 minute drive. That’s a meaningful friction reduction when you’re navigating SBIR applications, STRATFI transitions, or Other Transaction Authority contracts.

What DC does fundamentally lack though is a startup culture. There’s a reluctance to move fast, accept the unknowns and frankly, fail at all. The region’s talent has historically been socialized into the contractor mindset: stable, process-driven, security-conscious. We believe that’s changing, the Pentagon has issued statements saying, in fact, that that must change, but it’s been relatively slow thus far.

Austin, TX:

Austin’s defense tech story runs through one building. Here’s a list of defense acquisition arms that maintain a presence there.

T2COM (formerly known as Army Futures Command), Army Applications Laboratory, AFWERX, NavalX, Defense Innovation Unit, National Security Innovation Network and Joint Defense Innovation Unit.

This is the result of a deliberate DoD strategy, beginning when DIU established its first Austin outpost in 2016, to build a front door to the commercial tech ecosystem outside the Beltway. The primes are now following, hoping to keep their eyes on some of the novel technologies that the defense ecosystem has been boasting about down in Austin. In 2025, BAE Systems announced a $150 million campus expansion in Austin, adding 500 jobs. Oracle launched a defense ecosystem offering cloud infrastructure. The primes are there and there are sure to be more on their way.

Founders here are not socialized into the contractor mindset. The tech ecosystem is populated by builders who want to move fast, and the DoD innovation units that chose Austin specifically want to work with them. Austin’s strength is increasingly in AI, autonomous systems, cybersecurity, and drone technology.

San Diego, CA:

The maritime industry. Look, it’s being talked about daily.

The hubs above have legible narratives and established gravity. But the next tier is worth understanding and in some cases, there’s a structural tailwind that the broader market hasn’t fully priced in yet.

The Navy and Marine Corps have massive installations in San Diego, and that proximity shapes everything from who you can hire (veteran naval officers with domain expertise) to who will take your first meeting (program managers who work a few miles from your office).

The broader ecosystem includes legacy primes like General Atomics and Northrop, which have both been a significant employer of defense engineers in the region for decades, and a growing venture infrastructure that reflects the defense tech boom. Maritime defense, ocean tech, and unmanned surface and undersea vehicle development have found a natural home here.

Boston, MA:

What Boston has is something harder to replicate than the other hubs, and more durable in the long run: the deepest concentration of defense-relevant research institutions in the United States, generating a continuous supply of foundational IP and PhDs that the rest of the places listed in this article cannot match.

MIT Lincoln Laboratory has been the operational core since 1951. It is one of the most productive federally funded defense research institutions and has been the talent factory for the Boston defense ecosystem for seven decades. Since its founding, MIT/LL has spun off more than 100 startup companies.

Boston has become the center of gravity for autonomous maritime defense startups and it’s not just because of the robotics talent pool, though that matters. It’s physical access to operational testing water combined with the proximity to the Navy’s most important undersea warfare institution (NUWC Newport) in the country. HavocAI, a New North Ventures portfolio company currently has three vessel classes in active water testing in Rhode Island: the 38-foot Seahound, the 42-foot Kaikoa, and the 100-foot Atlas multi-mission vessel, all in the waters adjacent to NUWC Newport. That’s not coincidence, it’s deliberate proximity to the customer who will certify and procure the systems.

For founders seeking talent from the most legitimate research institutions in the country, or to partner with some of the most advanced laboratories in the country, there’s no better place. For founders seeking exposure to maritime testing infrastructure, there also may be no better place.

Here are a couple more that could be considered:

Huntsville, Alabama

This defense ecosystem was built up around the redstone arsenal and for decades it operated accordingly, anchoring itself in missile defense and operations. The current administration just announced plans for US Space Command to be headquartered there. That moves a ton of jobs and a Space focus over to Huntsville.

For a defense startup building in missile defense, directed energy, space domain awareness, or any other advanced missile technology, one could argue that there is now no better place in the country to be proximate to the customer.

Colorado Springs, CO

Known as a hub for military operations for some time now, Colorado Springs boasts itself with some serious hubs. USNORTHCOM, NORAD, Ft. Carson, US Space Combat Forces Command, USAFA, just to name a few highly regarded military entities in the Springs. Losing the US Space Command to Huntsville certainly stings but with the deep history that the Colorado Springs corridor has, defense tech, and the need for it aren’t going anywhere.

The Catalyst Campus operates as an accelerator with a direct pipeline to Space Force procurement. Between the young officers coming out of the Air Force Academy with real, genuinely innovative ideas, to the satellite operators and space domain awareness SMEs, this is an excellent opportunity to hire some real defense talent. Companies have and will continue to take advantage of that. The primes are there, with Northrop, Lockheed and L3Harris boasting massive operations but also, you’ve seen companies such as Parsons and Kratos beginning to thrive there as well, leading us to believe the smaller fish will and should be attracted to this area.

So here are a few aspects to consider when you’re looking at places to found your company… and of course whether you have the freedom and resources to relocate should be a primary consideration. But after that, what are you building? Who are you selling to? What stage are you at? Is there a way to similarly scale your technology to the commercial sector in these locations?

Exclusive-Pentagon spars with SpaceX over Starlink price hike during Iran war

The Pentagon and SpaceX are reportedly locked in a dispute over pricing for Starlink satellite communications services supporting U.S. military operations against Iran. SpaceX has sought higher pricing for bandwidth and terminal services being used in the Middle East, arguing that wartime operations create extraordinary demands on the constellation, including unpredictable bandwidth spikes, contested electromagnetic environments, and heightened risks to the satellites themselves. The Pentagon, meanwhile, argues that existing contracts should already account for surge military usage. The dispute highlights a growing tension in modern defense procurement: commercial infrastructure built for civilian markets is increasingly being relied upon for frontline military operations under active conflict conditions.

The broader implication extends far beyond Starlink. For years, the Pentagon embraced commercial technology providers to move faster and reduce costs compared to traditional defense contractors. That model works well in peacetime, but modern conflicts increasingly target the infrastructure layer itself, from satellites and cloud networks to GPS and subsea cables. For venture investors and defense startups, the lesson is becoming clearer: dual-use business models built around shared commercial infrastructure may struggle when national security missions require sovereign control, surge capacity, and resilience under attack. The companies most likely to win long-term defense contracts will be those architected from the start for contested operational environments where the infrastructure itself becomes part of the battle space.

More links to explore:

StarCube: New England Conference of Public Utilities Commissioners

Advanced nuclear is re-entering the New England energy conversation, and at the New England Conference of Public Utilities Commissioners, New North Ventures portfolio company StarCube represented the growing momentum behind small modular reactors and next-generation nuclear innovation in the region. Jeremy Hitchcock joined leaders from NextEra Energy Resources, GE Vernova Hitachi Nuclear Energy, Nuclear Innovation Alliance, and the Massachusetts Executive Office of Energy and Environmental Affairs to discuss how New England will need firm, scalable, carbon-free power to support electrification, AI infrastructure, and industrial growth. The discussion highlighted that SMRs and microreactors are rapidly moving from concept to deployable infrastructure, and that states creating clear regulatory pathways and long-term procurement certainty will be best positioned to attract capital, talent, and energy deployment.

Thanks for reading Securing Our Future! Subscribe for free to receive new posts and support our work.

Subscribe now

Radio Frequency: The Invisible Infrastructure Behind Modern Warfare

Wed, 20 May 2026 10:03:22 GMT

A blue ocean is forming in RF native infrastructure. The market is stepping down cleanly from institutional validation into early stage opportunity: Hidden Level has become the Series C anchor for passive RF airspace intelligence with its $65M Series C; MatrixSpace and Fortem define the Series B layer with deployable AI radar and counter-UAS infrastructure; and the Series A layer is filling in with companies like AnySignal, Distributed Spectrum, and CX2 across RF infrastructure, signal intelligence, and electronic warfare. The map below lays out that funding stage progression, from Series C validation down to seed stage wedges.

What makes the category interesting is that this is not just “counter-drone.” It is the emergence of a broader RF stack for sensing, spectrum awareness, airspace security, electronic warfare, satellite communications, and critical infrastructure protection. Passive RF collection and airspace awareness are becoming the arteries of modern dual-use infrastructure, with applications across defense, airports, stadiums, power grids, ports, autonomous systems, and space. The deal activity reflects that shift: $20M into MatrixSpace, $25M from Lockheed Martin into Fortem, $24M into AnySignal, $25M into Distributed Spectrum, and $31M into CX2 all point toward a market that is moving from point sensors toward programmable electromagnetic infrastructure.

The company that stands out at the seed layer is Nullspace. Unlike many companies on the map, Nullspace is not another passive RF collector or counter UAS platform. It is building the simulation and design software layer that could sit underneath the entire category. If Hidden Level and its peers represent the visible arteries of RF awareness, Nullspace is closer to the veins and design tissue that let future RF systems be modeled, optimized, programmed, and projected into complex electromagnetic environments. That makes Nullspace interesting as a potential “RF design foundry” for the next generation of antennas, arrays, radar systems, EW payloads, communications networks, and quantum hardware.

NY Firm Opens RF Testing Chamber to Boost Airspace Solutions Production

RF sensing company Hidden Level opened a new state-of-the-art anechoic chamber at its manufacturing campus in East Syracuse, New York, to accelerate the testing and production of its airspace security and counter drone systems. The chamber creates a shielded environment that blocks outside electromagnetic interference and absorbs internal signal reflections, allowing engineers to conduct highly precise, repeatable testing of RF sensing systems used for low-altitude awareness and C-UAS missions. By bringing testing capabilities fully in house, Hidden Level says it can move faster on product development, reduce reliance on third party facilities, and more quickly deliver validated systems to the US military and allied partners. The expansion also reflects broader investment in domestic defense manufacturing and RF infrastructure in Central New York, supported by partnerships with state economic development organizations and manufacturing programs, as the company continues scaling production to meet rising global demand for drone detection and airspace security technologies.

More links to explore:

Distributed autonomous platform for RF sensing expanded by Aerovironment

Quantum Systems acquires Estonia’s SensusQ

German drone manufacturer Quantum Systems has acquired Estonian AI driven intelligence platform SensusQ, bringing together advanced drone hardware and battlefield intelligence software into a more integrated defense technology stack. SensusQ’s team and operations in Estonia will remain in place, with plans to continue hiring engineers locally. The acquisition follows Quantum Systems’ broader expansion strategy after its major Series C raise in late 2025 and reflects growing demand for combining autonomous ISR platforms with AI powered intelligence analysis tools.

SensusQ, founded in 2020, developed software that fuses data from drones, cyber events, OSINT, sensors, and human intelligence into a unified operational picture for defense and security users. In 2024, New North Ventures participated in SensusQ’s seed financing round, helping support the company’s international growth and product expansion. The company had recently gone through financial restructuring in Estonia before the acquisition, but emerged with renewed momentum as demand for AI-enabled defense software continues to accelerate across Europe and NATO-aligned markets

As Deepfake Content Increases, Companies Fight Back

As deepfake content becomes more sophisticated and widespread, companies are racing to develop tools that can detect and defend against AI generated misinformation, impersonation, and fraud. Deepfakes have exploded from roughly 500,000 examples in 2023 to an estimated 8 million in 2025, creating growing risks for enterprises, governments, and online platforms. Modern deepfake detection now goes beyond text analysis and instead relies on multimodal systems that analyze inconsistencies across video, audio, images, metadata, and behavioral patterns to identify manipulated content.

One of the companies featured is Reality Defender, a New York based deepfake detection platform and portfolio company of New North Ventures. Founded in 2021, the company provides enterprise grade tools to identify synthetic media across audio, video, images, and text through APIs and web applications designed for governments, financial institutions, and large enterprises. Reality Defender previously raised significant venture funding and has continued expanding partnerships and deployments as demand for AI-driven fraud prevention accelerates globally.

Thanks for reading Securing Our Future! Subscribe for free to receive new posts and support our work.

Subscribe now

From Artemis to Acquisition: Securing the Earth–Moon Economy

Wed, 13 May 2026 10:03:12 GMT

Increased Demand For Innovation In Cislunar Capabilities

Last month, a crew of four astronauts successfully completed a historic journey around the Moon and back. Marking the first time humans have ventured that deep into space since 1972’s Apollo 17 mission, the flight was a true demonstration of America’s bold Space program. On April 1, Reid Wiseman, Victor Glover, Christina Koch, and Jeremy Hansen of the CSA launched from Kennedy Space Center, eventually reaching a record breaking distance of 252,756 miles from Earth and passing within 4,067 miles of the lunar surface. By April 10, they safely splashed down in the Pacific, returning in such great health that all four were able to walk off the recovery ship unassisted.

This was amazing for so many reasons, 54 years later “Still got it” America’s Space Program says like a former college basketball player says when draining a 3. From a defense lens, what’s so interesting is that the U.S. Space Force announced it was standing up a dedicated office to figure out how the military fits into whatever comes next in the Earth-Moon system. The juxtaposition between NASA proving humans can survive the journey, and the Space Force immediately asking who secures it is the story of cislunar right now.

On April 15, at the annual Space Symposium in Colorado Springs, Maj. Gen. Stephen Purdy, the senior advisor to the Secretary of the Air Force for space acquisition, announced the creation of the Cislunar Coordination Office, an acquisition task force whose mission will be to build the roadmaps for acquiring cislunar capabilities. Purdy made explicitly clear that this wouldn’t be an exploratory working group, this would be a collaborative effort to ensure we build out a roadmap for cislunar activity and then begin coordinating with industry.

It’s not surprising that we hear the Department of War going in this direction after President Trump’s executive order Ensuring American Space Superiority in December of 2025. The direction from this executive order establishes that America will have begun constructing the “initial elements of a lunar outpost by 2030” and thereby extends the DOW’s defined area of responsibility through to lunar space.

Although the executive order does not explicitly call for a military base on or around the Moon. What it does is mandate a sustained U.S. architecture, to include transportation, communications, and navigation, which by our estimation, inherently carries security implications.

Here is what most coverage of the Space Force’s cislunar ambitions breezes past: cislunar space is technically challenging in ways that GEO (Geosynchronous Orbit) and LEO (Lower Earth Orbit) are not, and the gap between the Space Force’s current toolset and what cislunar demands is significant. The fundamental issue lies in that the traditional orbital mechanics and toolsets can’t be used in cislunar like they can be in GEO and below. The moon’s gravitational influence starts to play a role once you get to the cislunar regime. The commercial space industry will be needed, and just like the Defense Industrial Base has been tasked with in so many ways before, a new blueprint will need to be created. What companies will dare to be part of such a challenging problem? What companies will dare to deal with such a vast area of responsibility?

Although Artemis II wasn’t a military mission, it changed the overarching strategic calculus by the DOW and specifically the USSF. For five decades, the Moon was inert from a national security standpoint, sure it was scientifically interesting, but militarily, it was irrelevant. As habitats start to form, infrastructure starts to be built, communication and power systems start to exist on or near the moon, there will become a need to monitor, protect, and in a contested scenario, defend them, and that is inherently a Space Force problem.

The Cislunar Coordination Office’s stated intent to partner with industry is significant, and the timing, coming immediately after Artemis II and within months of the December executive order, suggests the government is serious about moving from planning to procurement. They’ll define the capability gaps. They likely won’t be abstract impossibilities, but rather, well-defined and acquirable technology areas.

For investors paying attention to the defense-space intersection, the pattern here rhymes strongly with how SDA developed at proliferated LEO. Early government investment in domain awareness created the acquisition architecture that commercial operators eventually fed into. The companies building cislunar relevant capabilities now are positioning early for what will likely become a meaningful acquisition category by the late 2020s.

The Space Force has 9,000 Guardians. The entire organization could fit in a high school football stadium. It cannot build cislunar alone. The question is not whether the Space Force will be in the Earth-Moon theater, it will, it’s surely established that. The question is whether the companies of the current Defense Industrial Base will move fast enough to matter, because others will and we’ll be on the lookout for them!

Amazon vs. SpaceX: The High-Stakes Battle for Space Internet | WSJ

The race to dominate low Earth orbit (LEO) satellite internet is accelerating, with Amazon attempting to close the massive gap between its satellite initiative and SpaceX’s rapidly expanding Starlink network. Amazon has launched roughly 300 satellites so far for its LEO business, despite being authorized to deploy more than 7,000, while SpaceX has already launched approximately 12,000 Starlink satellites into orbit. LEO satellites, generally operating between about 100 and 1,200 miles above Earth, offer major performance advantages because signals travel shorter distances, resulting in faster internet connectivity for users on the ground. The video also notes that major space assets including the International Space Station and the Hubble Space Telescope operate in LEO, underscoring how strategically important this orbital region has become for both commercial and government activity.

The market opportunity for global broadband connectivity is enormous, with a SpaceX executive previously estimating the addressable market at roughly $1 trillion annually. Still, scaling a LEO network comes with major challenges. Unlike traditional geostationary satellites positioned more than 22,000 miles from Earth, LEO systems require thousands of satellites to provide comparable global coverage. Another challenge is that much of the Earth’s surface is ocean, limiting the number of traditional household customers directly beneath satellites at any given time. As a result, operators are aggressively targeting maritime and aviation customers including container ships, cruise lines, and airlines to maximize network utilization. Investors continue to see strong upside in the sector, with Goldman Sachs projecting the LEO satellite market could grow to approximately $108 billion by 2035, while Starlink is expected to become a major driver of SpaceX’s valuation and anticipated IPO plans.

Tech giants race to build space data centers to power AI boom

As AI demand pushes terrestrial power grids to their limits, major technology companies are increasingly looking to space as the next frontier for compute infrastructure. The article highlights how SpaceX, Google, Nvidia, Amazon, and Meta are exploring space based data centers powered by solar energy. Elon Musk recently predicted orbital data centers could become viable within “two to three years,” using networks of low Earth orbit satellites equipped with GPUs and connected via laser communications to operate as one massive AI compute cluster. The appeal is compelling: solar generation in space can be five to eight times more efficient than on Earth, while the near freezing conditions of space could dramatically reduce cooling costs for energy intensive AI servers.

Early efforts are already underway. Nvidia backed startup Starcloud successfully launched a satellite carrying Nvidia H100 GPUs that is already running Google’s Gemma AI model in orbit, and the company now plans to build a massive 5 gigawatt orbital data center. Meanwhile, Google is developing its “Project SunCatcher” satellite compute network, while Meta recently signed a deal to source up to 1 gigawatt of space generated solar power for AI infrastructure. The urgency is clear: the International Energy Agency projects global data center electricity consumption could reach 950 terawatt hours annually by 2030, equivalent to the output of roughly 120 nuclear power plants. While launch costs, radiation exposure, and space debris remain major hurdles, the article frames orbital compute infrastructure as a serious long term solution to the growing AI energy crisis.

More links to explore:

Havoc Raises $100M Series A to Power the Future of All-Domain Collaborative Autonomy

Havoc AI, a portfolio company of New North Ventures, announced a $100 million Series A round to accelerate the development of its all domain collaborative autonomy platform across maritime, air, and ground systems. The raise brings the company’s total funding to roughly $200 million since 2024 and included participation from major defense and technology investors including Lockheed Martin and SAIC. Havoc’s software enables a single operator to coordinate large fleets of autonomous systems operating together in contested environments, reflecting the growing defense focus on scalable, software defined autonomy.

The company says its autonomy stack already operates across more than 100 platforms and has accumulated over 25,000 hours of autonomous testing and deployments, generating more than 200 billion data points in the process. Havoc has gained particular traction in maritime autonomy through its Rampage autonomous surface vessels, while expanding into broader all domain operations spanning sea, air, and land. The funding comes amid rapidly growing investor and government interest in collaborative autonomy technologies designed to support distributed military operations and resilient logistics in contested environments.

HawkEye 360 Raises $416M in IPO (NYSE: HAWK)

HawkEye 360 raised $416 million in its IPO after pricing 16 million shares at $26 each, the top end of its expected range, giving the company an initial valuation of roughly $2.4 billion. The Virginia based company, which trades under the ticker HAWK, operates more than 30 satellites that collect and analyze radio frequency emissions globally for defense, intelligence, and national security customers. The IPO reflects growing investor appetite for space and defense technology companies, particularly those tied to signals intelligence and geopolitical monitoring.

The company’s stock surged roughly 30% on its NYSE debut, pushing its market value above $3 billion and signaling strong momentum for the broader defense tech and commercial space sectors. HawkEye reported approximately $118 million in 2025 revenue, achieved profitability for the first time, and disclosed a backlog of more than $300 million. Much of its business comes from U.S. government and allied nation contracts, including work supporting maritime tracking, GPS jamming detection, and RF intelligence collection in contested environments. The successful offering is also being viewed as an important bellwether ahead of the highly anticipated SpaceX IPO expected later this year.

Thanks for reading Securing Our Future! Subscribe for free to receive new posts and support our work.

Subscribe now