AutoWarCom And The 74 Billion Drone Stock Opportunity

I kept coming back to one number after I finished the briefing notes on my desk. Seventy-four billion. Not spread across a decade of vague modernization slides, but proposed for drones and the systems meant to stop them. That kind of figure changes how a market talks about a theme. It also changes how a factory floor has to be built. If you have spent any time around defense names, you already know the pattern: a speech, a budget line, a rush of headlines, then a long wait while contracts get written in language only lawyers enjoy. This cycle feels different, and I will explain why without pretending anyone has a crystal ball.

The proposal sitting in front of planners is simple to describe and brutal to execute. Stand up a four-star combatant command for autonomous warfare, nicknamed AutoWarCom in the notes circulating around the capital, and push it toward operational status in the fall of next year. Pair that organizational shove with a request to triple spending on autonomous warfare. The stated goal is stockpiling across drone categories, robotics, and the software that stitches them together, while building a supply chain that does not depend on a rival power for motors, blades, sensors, or the boring parts nobody photographs.

Why This Rearmament Cycle Feels Different From The Last One

Wars used to move slower than procurement. That sentence sounds almost nostalgic now. Ukraine showed what cheap one-way attack aircraft can do to expensive armor. The Gulf region then showed that the same idea can reach infrastructure people assumed sat outside the fight, including sites tied to data and energy. I am not interested in the cable-news version of those events. I am interested in the purchasing response, because that is what shows up in order books.

A senior defense official described the effort, in a speech at a Marine base in Virginia, as the fastest peacetime shift in modern military history. Bold line. Maybe too bold. Peacetime shifts have a habit of slowing down once committees arrive. Still, the organizational piece matters. A dedicated command with a four-star at the top is not a working group. It is a signal that autonomous systems are supposed to stop living as side projects inside older services and start competing for real money, real ranges, and real production slots.

Once sustained budget meets changed doctrine, the side that industrializes faster usually writes the next chapter. Software helps. Factories decide.

Field note from a former operator turned buyer

Owen West, a former Marine who also spent time as an assistant defense secretary and on a trading desk, is slated to lead the early effort alongside a Navy SEAL test pilot, Max Strasiser. West has been tied to the Pentagon innovation shop. His public line is worth sitting with: apply sustained budget to changed doctrine and you outperform; snap in artificial intelligence and you pull ahead because domestic labs still lead that race. I buy the first half more easily than the second. Budget and doctrine are levers a government can pull. Leadership in software is real, and it is also rented every quarter from chip supply, power, and talent that other industries are bidding for.

The March Inflection Nobody Budgeted For Cleanly

People inside the building have been circling March as the month the argument stopped being theoretical. That is when the gap between demonstration videos and wartime consumption became impossible to shrug off. Attritable aircraft, the polite term for machines you expect to lose, stopped being a niche presentation and started looking like a magazine you have to refill. Counter-drone gear stopped being a base-protection afterthought and started looking like a utility, closer to fences and generators than to a marquee fighter program.

Here is the part retail investors often skip. A theme can be correct and a stock can still be the wrong vehicle. I have watched that movie in cybersecurity, in space launch, and in every “picks and shovels” story that forgot the shovel company had one customer and a delayed invoice. The drone trade will rhyme with those episodes. Some names will compound because they own a bottleneck. Others will dilute, miss a test event, or discover that a services contract does not scale like a product.

What Seventy-Four Billion Actually Tries To Buy

The figure attached to the largest budget request on record for this slice of the portfolio is 74 billion dollars for drone and counter-drone technology. Treat that as a request, not a check. Requests get marked up, split across accounts, and sometimes delayed by the same politics that produced them. Even a haircut version would still be a step-change versus the old trickle of prototype money.

Spend tends to cluster. From what program managers describe, the buckets look roughly like this.

• One-way attack aircraft in volumes that make artisan assembly a joke
• Group 2 and Group 3 systems for reconnaissance and targeting, plus the larger Group 5 airframes that can linger
• Counter-UAS sensors, jammers, interceptors, and the command software that has to choose among them in seconds
• Ground robotics and the autonomous logistics that keep a dispersed force fed
• Domestic or allied components so a single export rule cannot freeze a line

The last bullet is the quiet one, and in my view the most investable if you are patient. Airframes get the photos. Motors, propellers, radio boards, inertial sensors, and battery management get the margin if a program standardizes on them. A former buyer put it to me in plainer words: you cannot tweet your way to a million engines.

Stockpiles, Not Demos

Doctrine is shifting from exquisite and few toward affordable and many. That does not retire the expensive aircraft. It changes the ratio. A lower-cost Group 5 platform that can take some intelligence, surveillance, and reconnaissance load off a legacy hunter-killer drone is exactly the sort of complement planners like, because it frees the costly airframe for missions that still need it. Recent contract chatter around a platform called ULTRA, tied to Ondas, fits that sketch: a reported near-doubling of a deal value, framed as a cheaper Group 5 option that could sit beside the Reaper rather than replace it overnight.

I would not build a portfolio on one contract revision. I would treat it as a tell. When a department nearly doubles a line item for a lower-cost airframe, it is usually testing whether industry can deliver quantity without the old cost spiral. If the test works, follow-on orders are where the equity story either becomes real or fades into a pilot that never left the range.

Conflict-Free Supply Is The Actual Constraint

You can announce a command in a season. You cannot announce a motor plant into existence. The public line from the department is clear enough: engines, blades, sensors, and related parts should come from domestic lines or from places that will not become a conflict problem. That is a supply-chain project wearing a uniform.

China still sits in the middle of a lot of small electric propulsion, magnets, and circuit-board capacity. Unwinding that is not a press release. It is capex, workforce, testing, and a willingness to pay more per unit for a while. Companies that already machine, wind, or assemble outside that dependency have a head start. Companies that relabel an imported kit will get found out the first time a compliance team opens a crate.

Perhaps the most interesting aspect, if you follow industrial names rather than pure software, is how unglamorous the winners may look. A specialist composites shop. A radio firm that already sells to ranges. A battery pack assembler with the right paperwork. The market likes a ticker with “autonomy” in the deck. The invoice often goes to the firm that shipped on Tuesday.

Data Centers Changed The Customer List

A thread I have been tracking since late January is awkward for people who still file drones under “military only.” Every large data campus is a soft target if the threat is a cheap aircraft with a small warhead and a decent navigation kit. When sites in the Gulf were hit, the hypothetical became a facilities problem. Power, cooling, and fiber already keep operators awake. Kinetic intercept just joined that list.

That widens the buyer set. Base protection budgets are one pool. Critical infrastructure, energy terminals, and private campuses are another, slower pool, gated by insurance, liability, and rules about who may jam or shoot. I have found that investors who only model the defense line miss the second pool, and investors who only model the second pool forget how long civil approvals take. Both clocks matter.

1. Map the threat a site actually faces, not the threat in a keynote
2. Pick a sensor layer that can live with existing security staff
3. Add an interceptor that is legal, affordable, and reloadable
4. Write the playbook before the first incident, not during it

None of that is a stock tip. It is a demand sketch. Demand sketches get you direction. Filings get you position size.

How A Four-Star Command Changes The Buying Rhythm

Combatant commands do not design every gadget. They write requirements, set priorities, and pull services toward a common picture. If AutoWarCom becomes real on the stated timeline, the fall of next year is the organizational start, not the revenue start. Money has to be appropriated, programs have to be named, and test events have to stop being theater.

The useful investor habit is to separate three clocks.

Clock one: speeches and org charts
Clock two: budget lines that survive markup
Clock three: hardware accepted into inventory

Equity markets trade clock one. Cash flow lives on clock three. The gap between them is where drawdowns happen, even in a theme that is directionally right. I have sat through enough of those gaps to prefer companies that can fund the wait with existing revenue rather than with a single heroic award.

Where Ondas Sits In The Conversation

Market notes over the past weekend pointed to Ondas after a reported increase in the value of work tied to its ULTRA platform. The framing is a lower-cost Group 5 drone that could complement a Reaper-class aircraft and, over time, take some surveillance missions. That is a plausible niche. It is not a monopoly, and it is not a guarantee that follow-on volume arrives on the schedule a shareholder wants.

What I like about the setup, as a story rather than as advice, is the fit with the stated doctrine. Planners want mass and they want to stop burning the most expensive hours on routine orbits. A platform that is good enough, cheaper to lose, and fed by parts that can be sourced outside a rival supply chain matches the brief. What I do not like is how quickly a small-cap defense name can gap on a paragraph and then spend two quarters explaining milestones. Position size is the whole game in that neighborhood.

Follow the purchase orders, not the podium. A doubled pilot is a clue. A multi-year production lot is a business.

The Rest Of The Field Is Not Sitting Still

Prime contractors will not hand this lane to newcomers out of courtesy. They already own ranges, security clearances, and the relationships that turn a prototype into a program of record. Expect them to partner, acquire, or quietly replicate. Specialist firms still have room where a prime is slow: a novel airframe, a counter-drone effector that is actually cheap, a radio that works when everyone else is jamming.

Public names across the drone and counter-drone complex will all claim the same tailwind. Your job is narrower. Who has hardware in inventory today? Who relies on a foreign subcomponent that a new sourcing rule would ban? Who recognizes revenue on delivery rather than on a slide? Those questions sound basic because they are. They also filter out a surprising amount of noise.

Artificial Intelligence Is The Slogan And The Bottleneck

West’s line about snapping in AI is the sentence every deck will quote. Autonomy that works in a cluttered electromagnetic environment is harder than a model demo on a clean dataset. Targeting support, swarm deconfliction, and automatic threat ranking all need compute, trusted data, and a human still accountable when the call is wrong. The firms that already run that loop on a test range have an advantage over firms that licensed a model last spring and printed a new cover.

There is a second, less discussed constraint. Power and chips are spoken for by commercial data centers. A defense program that assumes unlimited inference at the edge will learn about lead times the hard way. I would rather see a company that designs for smaller models and degraded communications than one that needs a perfect link to a cloud it does not control.

Installation Defense Is A Repeating Budget, If It Sticks

Fortifying sites around the world sounds expansive until you price it. Sensors on a fence line, a local operations cell, reloadable interceptors, and training for people who did not sign up to be air defenders. If that package becomes a standard line in base budgets, the revenue is closer to a utility. If it stays a supplemental response after each headline, the revenue is lumpy and the multiples should be too.

Private operators of data campuses face a cousin of the same choice. Some will buy a service. Some will demand that landlords handle it. A few will overbuy after an incident and then cut the contract when the news cycle moves. Recurring beats heroic. I keep writing that because the alternative is how people overpay for a theme.

A Practical Way To Read The Theme Without Falling For It

None of this is a recommendation to buy or sell any security. It is a map of where reported policy and reported contracts are pointing. If you underwrite the theme, I would pressure-test four items before the story gets louder.

• Appropriation versus request. A proposal is not cash until it survives the process.
• Production evidence. Photos of a prototype are cheaper than a second shift.
• Parts origin. Domestic or allied content has to be documented, not asserted.
• Customer concentration. One program office can pause you. A mix of defense and infrastructure work is sturdier, and harder to build.

Valuation is the fifth item, and it is the one people skip when a command gets a catchy name. Small defense suppliers can trade at prices that assume the entire budget lands in their hangar. It will not. Even a successful rearmament spreads across primes, specialists, foreign partners, and classified lines you will never model cleanly.

Risks That Can Dull A Correct Theme

Budgets get cut. Ceasefires change urgency without deleting the lesson. Export rules tighten and then loosen. A test failure on a live range can freeze a line for a year. Competition from allies, who are running their own volume programs, can cap price. And the plainest risk: the department says autonomous and then spends the money on a familiar airframe with a new sticker.

There is also execution risk on the command itself. Standing up a four-star headquarters by next fall is an administrative sprint. Headquarters do not build drones. They can, however, slow the services if the seams are messy. I have watched joint offices add process faster than they add output. Worth watching, not worth romanticizing.

What I Will Be Watching Into Next Fall

First, whether the autonomous warfare request stays near the stated scale after markup. Second, whether component sourcing rules show up as contract clauses rather than speeches. Third, whether lower-cost Group 5 work moves from a revised award to a production tempo. Fourth, whether counter-drone packages start appearing as standard installation kits instead of emergency buys. Fifth, whether any of the specialist names can show backlog that is not a single press release.

If those boxes tick, the rearmament supercycle people have been sketching stops being a slogan. If they do not, you still learned something useful: the doctrine changed faster than the factories, and the equity needed more time than the headline. That outcome is common. It is also survivable if you did not confuse a speech with a purchase order.

A Note On How To Hold The Idea

Themes like this reward people who can sit with boredom. The interesting week is the speech. The profitable years, if they come, look like hiring notices in a secondary city and a footnote in a 10-Q about units delivered. I would rather own that boredom at a sensible price than own the excitement at any price.

You do not need to predict the exact shape of AutoWarCom to take the industrial point seriously. Mass, cheaper hours in the air, installation defense, and parts that can be sourced without a rival’s permission are the four legs. Companies that can prove two of them with paperwork are more interesting than companies that can narrate all four. That is the filter I keep taped above the monitor, and it has saved me from more stories than it has cost me.

Filter: budget that survives + parts you can source + units you can count

The war changed shape. The budget is trying to catch up. Whether a given ticker catches up with the budget is a separate question, and it is the only one that pays.