Cost per shot is a laser's cheapest number

I have written here before about why a four-million-dollar interceptor missile is a losing trade against a cheap drone, and about how high-power microwaves defeat a drone that no jammer can touch. The laser is the third member of that family, and it gets the friendliest press of the three, because it comes with a number that sounds like a solved problem.

The number is real. It is also, read carefully, a statement about electricity and nothing else.

Where the dollar comes from

The Congressional Research Service states the cost-per-shot claim precisely, and the precision is the whole point. In its standing report on directed energy weapons for Congress, it writes that high-energy lasers might offer lower logistical requirements, lower costs per shot and deeper magazines than conventional munitions, and then attaches a condition in the middle of the sentence:

In general, HEL weapons might offer lower logistical requirements, lower costs per shot, and—assuming access to a sufficient power supply—deeper magazines compared with traditional munitions.

And in a parenthesis immediately after, it says where the dollar figure comes from:

Although a number of different types of HELs exist, many of the United States' current programs are solid state lasers, which are fueled by electrical power. As a result, the cost per shot would be equivalent to the cost of the electrical power required to fire the shot.

That is an accounting identity, not a performance claim. A solid-state laser's ammunition is electricity, so its marginal cost per engagement is the electricity bill for a few seconds of firing. Nobody is lying when they quote it. But a marginal cost tells you nothing about the fixed cost, the availability, or whether the weapon works on the night you need it, and those are the three things a base commander is actually buying.

The physics bill, in the government's own words

The same CRS report carries an appendix listing the potential limitations of high-energy lasers. It is short, unclassified, free, and it is the most useful page in the laser-weapon literature. Four items in it matter for counter-drone work.

Line of sight. Laser light travels essentially straight, so the weapon can only engage what it can see. CRS notes this rules out over-the-horizon targets and targets hidden behind intervening objects, and that engagement ranges against low-flying targets are limited as a result. A drone in the terrain shadow of your own hangar is not a target.

Atmosphere. Water vapour is the main absorber, but CRS also lists sand, dust, salt particles, smoke and other air pollution, all of which absorb and scatter light, while turbulence defocuses the beam. Lasers can be tuned to wavelengths where water absorption is lower, and adaptive optics can partly correct turbulence in real time. The conclusion is still blunt:

Even so, lasers might not work well, or at all, in rain or fog, preventing lasers from being an all-weather solution.

Thermal blooming. This one is counterintuitive enough that it rarely survives into popular coverage, and it deserves to. A laser holding on one bearing heats the column of air it is passing through, and hot air defocuses the beam that made it hot. CRS:

This effect, called thermal blooming, can make lasers less effective for countering targets that are coming straight at them, on a constant bearing (i.e., "down-the-throat" shots). Most tests of laser systems have been against crossing targets rather than "down-the-throat" shots. In general, thermal blooming becomes more of a concern as the power of the laser beam increases.

Read that twice. The geometry that suffers most is the target flying directly at you, which is the geometry of an attack on the thing the laser is defending. And the effect gets worse as you scale power, which is the one lever the entire field has been pulling for a decade. The Department of Defense roadmap described in the same report aims to take high-energy laser weapons from around 150 kilowatts, described as currently feasible, to a 500 kilowatt class. Scaling power is the plan. Thermal blooming scales with it.

Saturation. CRS is specific about the tempo. A laser attacks one target at a time, requires several seconds to disable it, and requires several more to slew to the next. That puts a hard ceiling on how many targets one aperture can service in a given window, and the stated mitigation is to install more than one laser, up to the space and energy the platform has. Against a swarm, the magazine depth that sounded infinite is bounded by the clock instead of by rounds.

The Government Accountability Office reached the same place from the other direction in its 2023 science and technology spotlight on directed energy weapons, which lists among the challenges that these weapons "are generally less effective the farther they are from the target, and atmospheric conditions and cooling requirements can limit their effectiveness," offering fog and storms as the example. GAO also gives the clearest one-line statement of what a high-energy laser is: a beam of at least one kilowatt, "200,000 times greater than a typical laser pointer and is capable of melting steel," used on one target at a time.

Kilowatts on the label, watts per square centimetre on the target

The number in a laser's name is output power at the aperture. What kills the drone is irradiance on the target, energy delivered per unit area per unit time, held on one spot long enough to burn through it. Those two quantities are separated by every metre of atmosphere in between, and by how tightly the beam control system can hold a moving aim point.

In February 2024 the Army sent four Stryker-mounted 50-kilowatt laser prototypes, the Directed Energy Maneuver Short-Range Air Defense system, to the US Central Command region. Army Vice Chief of Staff Gen. James Mingus described the goal at the time as experimenting in a live environment with weather and dust storms, and then put the physics in a single question:

You may have a 50-kilowatt laser, [but] at 10 kilometers can you put at least four kilowatts in a centimeter square because … that's what you need to burn through a quarter inch steel plate?

He answered it himself:

But that's really hard to get … from a big beam to get the small portion of it on the exact spot to be able to burn at that high intensity and any kind of dust particle or that starts to disrupt that.

That is a four-star officer explaining, in public, that the headline kilowatt figure and the delivered intensity are different numbers, and that dust sits between them. It is the most honest sentence anyone in the directed-energy business has said on the record, and it came from the customer rather than the vendor.

What the soldiers said

Three months later the feedback arrived. Testifying to the Senate Armed Services airland subcommittee in May 2024, Army acquisition head Doug Bush said the initial response to the Stryker lasers was, in the reporting's summary, not overwhelmingly positive. His own words identified the problem as integration rather than optics:

That [50-kilowatt] power level is proving challenging to incorporate into a vehicle that has to move around constantly — the heat dissipation, the amount of electronics, kind of the wear and tear of a vehicle in a tactical environment versus a fixed site.

Heat dissipation. Electronics volume. Wear. None of those appear in a cost-per-shot calculation, and all three are recurring costs. Bush also noted a contrast worth keeping: the 20-kilowatt Palletized High Energy Laser, a lower-power system used at fixed sites, was proving successful for some setups. Less power, no vehicle, better result. That is not a paradox. It is what happens when you stop asking the weapon to also be a truck.

The budget moved with the feedback. The Army had planned to spend nearly $126 million on the 50-kilowatt effort in fiscal 2025 and requested $88 million instead, and reduced its projected fiscal 2026 to 2028 spending on the program by about $186 million against the previous year's plan. CRS records the same arc from the program side: a 2021 shoot-off at Fort Sill in which the system defeated drones but failed against mortar rounds, one competitor withdrawing, further testing in 2022 and 2023, then the Middle East deployment and the feedback.

The clean room

Of everything in this file, the detail I would put in front of anyone evaluating a laser for base defense is a remark by Lt. Gen. Sean Gainey, then head of Army Space and Missile Defense Command, in August 2024, about what maintaining one of these actually involves:

When I talk to sustainment, I'm talking the ability to do maintenance on laser systems and with dusts and dirt, you have to have a sterile environment right now.

The fix the Army was pursuing was not to make the optics dust-proof but to make the parts bigger, so that a technician replaces a whole module rather than opening an assembly in the field. As Gainey put it, instead of taking one smaller part out, maybe you replace that whole entire part so you do not have to take it into a clean room. A vendor in the same reporting confirmed the constraint from the industry side, saying that for some components, like the optics, you still need a clean room, while others had been redesigned for front-line service.

So: a weapon whose shot costs a dollar, whose optical train may need a controlled environment to service, mounted on a vehicle in a dust storm. Both facts are true at once. Only one of them is in the brochure.

Which is exactly why the Pentagon is now spending a year on it

Joint Interagency Task Force 401, the joint counter-drone organisation stood up in August 2025, is running a directed-energy pilot at five installations: Fort Huachuca in Arizona, Fort Bliss in Texas, Naval Base Kitsap in Washington, Grand Forks Air Force Base in North Dakota and Whiteman Air Force Base in Missouri. The systems are to sit on those bases for 365 days, run and maintained by service members rather than by contractor field service representatives, and a directed-energy shoot-off is scheduled for December at Dugway Proving Ground in Utah.

Brig. Gen. Matt Ross, who leads the task force, drew the line between the two questions as cleanly as anyone could ask. First, on what the pilot covers:

It's a directed energy pilot — and so, when we say "directed energy," we say it to be inclusive of all novel technologies — but what we're really focused on right now is high-energy lasers and then high-power microwaves.

Then, continuing in the same remark, the sentence this whole report is about:

And we have a number of systems, and we've proven that the science works — like you can defeat a drone with both a high-energy laser and then with a high-power microwave — but we need to figure out how to use them successfully in an operational context.

The stated objectives of the pilot are power requirements, cost burdens, repair tempos and operational tactics. Ross explained why he needs them, and it is the cost-per-shot problem restated as a procurement problem:

It's only with that type of information that I'll be able to go back to the services and say, 'I know you want to integrate directed energy — here's what it costs … so you can do a relative comparison of that type of defeat mechanism or mitigation, compared to a more conventional defeat mechanism where you're just going to sense and maybe use interceptors,'

He also said the goal is to develop tactics, techniques and procedures for using them on a base in a way that does not impede the critical functions of that base, which is a polite way of saying that a weapon which cannot be fired without shutting the airfield has a cost too.

Five sites across Arizona desert, west Texas, Puget Sound, the northern plains and Missouri is not a random sample. It is a weather sample, and a humidity sample, and a dust sample. The task force has designed the experiment around exactly the variable this report is about.

Why this beat cares

My first career was counter-IED and electronic warfare, which is a trade built on the difference between a system's specification and its behaviour on a road at the wrong time of year. Jammers have a cost-per-shot story too: the power is free once the vehicle is running. That framing was never the useful one. The useful questions were how much of the spectrum the thing covered, what it did to my own radios, how long the amplifiers lasted in heat, and whether the crew could keep it working without a depot. I have written about the second of those, electronic fratricide, before.

Lasers are in the same category of technology and deserve the same category of question. I want them to work. A directed-energy system that reliably defeats group 1 to 3 drones at a fixed site changes the economics of base defense in a way nothing else does, and the twenty-kilowatt fixed-site results suggest that narrower version of the promise is closer than the vehicle-mounted version. None of the limitations above says lasers do not work. They say lasers are a system, not a shot, and the system is where the money and the risk live.

The cheap-shot framing is dangerous mainly because of what it invites you to skip. If the shot is free, you stop asking about the generator, the chiller, the optics technician, the visibility minimum, and the number of simultaneous targets the aperture can service before the fourth one arrives. Those are the questions the next year of pilot data is going to answer, on five bases, in five climates, with soldiers rather than engineers holding the controls.

What I could not confirm

I did not open the JIATF-401 pilot announcement or any task force document directly. The Department's media server has refused every automated request I have made to it over several weeks, and the war.gov and army.mil pages for this story also refuse. Everything in the pilot section comes from a single reporter's account of an on-the-record media roundtable, which I opened and read in full, and which carries the quotations attributed to Brig. Gen. Ross. It is careful, named-source reporting from a specialist outlet, and I have quoted only what it puts inside quotation marks. It is not a primary document, and I have not seen one.

The soldier-feedback material is likewise press reporting of congressional testimony, not the testimony transcript. I have quoted Bush and Mingus from the article that reported them, and the "not overwhelmingly positive" characterisation is that reporter's summary rather than a quotation from Bush, which is why it appears here as a summary. The same characterisation is reproduced in the CRS report, which cites the same article, so the two are not independent confirmations of each other.

The CRS report I have leaned on hardest is the version dated 11 July 2024, which is the most recent one carried by the public mirror I can reach. Congress.gov refuses automated requests, so I cannot rule out a newer revision with different figures. Its funding table runs only to fiscal 2025. The GAO spotlight is from May 2023. Directed energy moves quickly, and in a fast-moving field a two-year-old assessment of maturity is a floor, not a snapshot.

I have not quantified anything myself. There is no calculation in this report: no dwell time, no attenuation coefficient, no engagement ranking. Beam propagation is modelled with real physics by people who do it for a living, and the numbers depend on wavelength, aperture, range, turbulence and target material in ways a report like this cannot shortcut. My published research is in microwave spectroscopy, and none of it involves laser weapons. I have never operated one.

Finally, nothing here is a product assessment. I name programs and companies only where the cited sources do, and no system in this report is evaluated, ranked or recommended.

The signal

A laser weapon's cost per shot is a genuine advantage and a genuinely small number, and it describes one line of the invoice. The other lines are prime power, thermal management, optics maintenance under field conditions, availability in weather, and how many targets a single aperture can service before the raid is through.

The Pentagon knows this. That is what a 365-day pilot with service members doing the maintenance is for, and it is why the task force commander separated proving the science from figuring out the operational context. When someone quotes you a dollar a shot, the correct follow-up is not skepticism about the dollar. It is: a dollar a shot, in what visibility, against how many at once, and how long is it down when the optics need cleaning.