Very interesting speculation. For what it's worth, here's my comment. You said "Another form of redundancy is decoys. If you launch a hundred decoy drones for every armed drone, the enemy will waste effort shooting down decoys." This reminds me of one of the steps in ballistic missile development. American designers thought it would be useful to add decoy warheads to missiles, in order to frustrate anti-ballistic missile defenses. In order to make the decoys work, they would have to mimic real warheads - similar size, mass, etc. But, if you were going to make something that looked like a warhead and had the same mass as a warhead, you might as well make it a warhead. Thus was born the MIRV - Multiple Independently targeted Re-entry Vehicles. In the same way, I expect that it won't be cost-effective to build decoy drones - just build more and more.
Nothing says you can't have a mix of drone types in a swarm. Some of them could be anti-drone drones, some could be drones that kamikaze on radars etc....
One of the things I would note is that drone motherships carrying other drones is absolutely a thing. As is the potential at least for drones to be launched from something apparently innocuous like a shipping container. If Israel wanted to attack Iran with drones the obvious thing to do would be to send a container ship to somewhere apparently innocuous like Mumbai and launch many many drones from it as it got close to Iran. A drone fishing boat that had drones in the holds where there used to be fish would also work for this kind of thing. That's maybe not as relevant for Israel vs Iran but it absolutely is something that countries in the vicinity of West Taiwan ought to be looking at
"An implication is that it would be really helpful to be able to jam or penetrate the guidance systems of an enemy’s small machines."
This is already a big deal, and has been for a long time. The Russians (and not just them) are very good at jamming and "electronic warfare" and achieving important battlefield affects using EM spectrum radiation. They've been using, selling, "advising", and giving away good GPS jammers everywhere the US has been operating for two decades.
More recently, the US has been giving lots of GPS-guided M982 Excalibur artillery shells to the Ukrainians. Early on in the war, these worked like a charm. Soon enough, effective GPS-jamming makes them go off course enough to be worse than dumb shells. You might get lucky with dumb shells and gradually dial them in. On the other hand, jammed guided-shells will try hard to go to the wrong spot and you can't adjust to make that better.
The people who designed the GPS system for military purposes were not dummies and anticipated this problem, and thus designed the system to be jam-resistant capable and transmit encrypted frequency-hopping signals with only US military GPS receivers getting frequently updated with the decryption keys.
But they made the mistake to only "go encrypted" in the event of a big war when GPS-jamming would become a problem. And so they just let the whole GPS system transmit in "plain text" for a long time, and then permitted global private civilian sale and use of GPS receivers. Which became cheaper and more and more widely adopted.
Until it became an indispensable public good upon which billions of people and large parts of the global economy had come to rely on a regular basis, and thus impossible to ever covert to exclusive, encrypted, and jam-resistant US military use ever again.
That was also recognized as an irreversible reality a long time ago, as was the need to develop and deploy a new, parallel jam-resistant geolocation system. As I understand it, this has been in the works for a long time, but not pursued with a sense of urgency, and so total switchover for military purposes still seems to be a ways out. Maybe recent developments will encourage people to speed it along.
I think the major theme here is asymmetry. Doing more at a lower cost is a warfare game-changer and has massive implications across the globe as well as on how we focus our financial resources in the US. Similarly, Cyber must be added to the asymmetry equation as it is the ultimate example of it. Cyber also has the added advantage/disadvantage of being hard/impossible to source (you can't watch the "cyber drones" slowly flying over from Iran). We urgently need to revisit our military priorities and spending in the US. In addition, we need to recognize that our historic advantages--natural resources, vastly superior technological lead, work ethic, education--may no longer exist and will require significant changes (cultural, political) to our society. This isn't just a military hardware issue.
I wonder how long it will be before we see functional Hunter-Seeker assassination drones? How will a politician defend against a few hundred of these insect sized drones with facial recognition?
The huge lesson contained in the Georges Clemenceau quote...... "Generals always prepare to fight the last war." (although possibly apocryphal) springs to mind here.
This is usually meant as a criticism, but "preparing to fight the last war" is kind of something you have to do anyway even if you know the next war is probably going to end up being different in important ways. This is because there almost always someone else out there with the ability to fight the same kind of war as the last one, and you don't want them to try it. The only thing dumber than preparing for a war like the previous one is *not* preparing for it, having to do it anyway, and then *losing*.
I don't think you've thought clearly about the range of small machines. Aircraft carriers are after all largely merely a way of transporting smaller machines closer to targets than they can fly on their own. Why cannot Israel use small machines against Iran if they use larger machines to transport them closer before launching them?
For the last two years, I have helped transfer non-lethal technology including drones to forces in Ukraine.
Lethal autonomous weapons are currently deployed in Ukraine in order to hit targets despite the jamming that prevents human control of radio-operated vehicles.
The components that support autonomy (sensors, compute and power) take up space and currently impose a floor on how small these devices can go. But they are small enough, because small is relative, and the machines do not have to be as small as insects to nevertheless create a new way of war.
Their size is relative to the capacity of defense systems to detect them. It also matters in a larger sense, because small devices can be made cheaply, and their numbers can overwhelm defense systems. They can overwhelm those systems in two ways: 1) they can target a lot sites, more than the enemy has expensive systems to cover. 2) they can target a single site with more attack vehicles than the enemy has defense missiles.
Size and cost are generally correlated, although the relationship is not 1:1. That is, by shrinking the size of devices, more can be made on the same budget. This logic doesn't strictly hold for Western militaries, where the incentives are perverse, but it does in Ukraine, because Ukraine is fighting a poor man's war. By driving down cost, one country can produce equipment in quantities that overwhelm the opponent, and also set up an advantageous trade of low-cost equipment for high-cost. All things being equal, this is how a war of attrition can be won.
Unfortunately, the US defense procurement process is spec-driven and its minimum timeline from need to deployment is measured in years, by which time most equipment has been obviated. What the US must do if it is to defend itself or its allies is create fast feedback loops between technologists and battlefield users in hot wars, short-circuiting the DOD's spec-driven culture in order to iterate more quickly and make what is both necessary and asymmetric to the enemy's capabilities. This is currently not how defense tech is produced.
Apparently, Russian officers know command soldiers in Ukraine via radio while monitoring them with drones in real time. It’s basically Warcraft at this point.
Depending on the theatre, if we consider the proposition at its scaled potential (i.e. mass swarms), then it becomes similar to an advance force ahead of occupying infantry. However, any infantry that occupies a territory will still be subject to mass counterattacks, likely from a stronger position when compared to the invading force. Strategically, you still need everything else - navy and air force for logistics and delivery of personnel, weapon systems, ammunition and resupply. To knock those systems out, you still need long range weapon strategic weapon systems.
Drones may offer a superior platform for localised dominance (eg over tank battalions), but not a significant strategic leap. Ukraine is seriously held back from using long range weapons, and we are still talking about small geographical distances. So they are forced into using drones.
In the Pacific, the most likely theatre against the CCP, you need strategic capability over vast distances. The next iteration is not likely drones, but hybrid space weapons or pure space deployed weapons. Countries use their non-defense industrial base advantages in total war - America's in the next war will be SpaceX, autonomy, software and AI. If you have cheap launch costs, you can have your war factories still position far away from the Pacific, launch vehicles into space, have them maneuver and deploy in a 30 minute window across the Pacific (example being the often discussed "Rods from God" concept [https://austinvernon.site/blog/starshipsuperweapon.html#:~:text=Rods%20from%20God&text=In%20the%201950s%2C%20a%20Boeing,feet%20underground%20to%20destroy%20bunkers.]
When you need to knock out factories, ports, refineries, rail, you need longer range deployment. Space is that channel. This also goes for directed energy weapons.
I can understand if you combine a Space based deployment with a cluster of drones that break apart from a single rod fired from space, and then swam locally to dominate a target. However, the hard part is the cheap reusable space deployment, not the drones.
So yes, drones will be valuable for smaller localised conflicts, but for the one that matters in the next thirty years - the Pacific - it will lend itself towards America's current advantages.
Conventional wisdom right now seems to be that the key feature of drones is "attritability." Losing a drone is not very bad, since they're cheap and you don't lose the skilled pilot. So you can make more and riskier attacks using drones than using manned systems.
My son has been working in the area of very small and light fully autonomous drones. Only by being autonomous can you avoid electromagnet interference with communications.
He is looking at an initial commercial markets using them to defend farms against birds (he lives in Napa where birds like grapes making this a real issue). His final market is imagined at defending against small drone swarm attacks with smaller/cheaper drones.
Very interesting speculation. For what it's worth, here's my comment. You said "Another form of redundancy is decoys. If you launch a hundred decoy drones for every armed drone, the enemy will waste effort shooting down decoys." This reminds me of one of the steps in ballistic missile development. American designers thought it would be useful to add decoy warheads to missiles, in order to frustrate anti-ballistic missile defenses. In order to make the decoys work, they would have to mimic real warheads - similar size, mass, etc. But, if you were going to make something that looked like a warhead and had the same mass as a warhead, you might as well make it a warhead. Thus was born the MIRV - Multiple Independently targeted Re-entry Vehicles. In the same way, I expect that it won't be cost-effective to build decoy drones - just build more and more.
That was my thought too, that the drone is probably more expensive than the C4 that makes it go boom.
Nothing says you can't have a mix of drone types in a swarm. Some of them could be anti-drone drones, some could be drones that kamikaze on radars etc....
A few could be stealthy while the majority light bright on radar.
Drones are absolutely the new thing.
I wrote a thing shortly after the Ukraine invasion started where I noted the importance of drones and earlier this year I wrote a follow up - https://ombreolivier.substack.com/p/attack-of-the-drones-revisited
One of the things I would note is that drone motherships carrying other drones is absolutely a thing. As is the potential at least for drones to be launched from something apparently innocuous like a shipping container. If Israel wanted to attack Iran with drones the obvious thing to do would be to send a container ship to somewhere apparently innocuous like Mumbai and launch many many drones from it as it got close to Iran. A drone fishing boat that had drones in the holds where there used to be fish would also work for this kind of thing. That's maybe not as relevant for Israel vs Iran but it absolutely is something that countries in the vicinity of West Taiwan ought to be looking at
"An implication is that it would be really helpful to be able to jam or penetrate the guidance systems of an enemy’s small machines."
This is already a big deal, and has been for a long time. The Russians (and not just them) are very good at jamming and "electronic warfare" and achieving important battlefield affects using EM spectrum radiation. They've been using, selling, "advising", and giving away good GPS jammers everywhere the US has been operating for two decades.
More recently, the US has been giving lots of GPS-guided M982 Excalibur artillery shells to the Ukrainians. Early on in the war, these worked like a charm. Soon enough, effective GPS-jamming makes them go off course enough to be worse than dumb shells. You might get lucky with dumb shells and gradually dial them in. On the other hand, jammed guided-shells will try hard to go to the wrong spot and you can't adjust to make that better.
The people who designed the GPS system for military purposes were not dummies and anticipated this problem, and thus designed the system to be jam-resistant capable and transmit encrypted frequency-hopping signals with only US military GPS receivers getting frequently updated with the decryption keys.
But they made the mistake to only "go encrypted" in the event of a big war when GPS-jamming would become a problem. And so they just let the whole GPS system transmit in "plain text" for a long time, and then permitted global private civilian sale and use of GPS receivers. Which became cheaper and more and more widely adopted.
Until it became an indispensable public good upon which billions of people and large parts of the global economy had come to rely on a regular basis, and thus impossible to ever covert to exclusive, encrypted, and jam-resistant US military use ever again.
That was also recognized as an irreversible reality a long time ago, as was the need to develop and deploy a new, parallel jam-resistant geolocation system. As I understand it, this has been in the works for a long time, but not pursued with a sense of urgency, and so total switchover for military purposes still seems to be a ways out. Maybe recent developments will encourage people to speed it along.
It’s happening!
Drone delivery is available in my neighborhood (DFW area of TX) from the local Wally Mart. Here’s what it looks like in action:
https://youtu.be/WdJ_vW5swec?feature=shared
Before you dismiss this as a novelty for the super lazy, just take a look at the possible use cases like ordering eggs:
https://youtu.be/7wuSjGzyXH0?si=7uiohagkwbB0eHD7
The US is not good at producing cheap defense products. 100,000 $400 small machines could well prove more effective than 1,000 $40,000 small machines.
Ukraine OTOH is absolutely making lots of the former. Hundreds and thousands of sub $1000 kamikaze drones that carry a grenade or anti tank warhead
I think the major theme here is asymmetry. Doing more at a lower cost is a warfare game-changer and has massive implications across the globe as well as on how we focus our financial resources in the US. Similarly, Cyber must be added to the asymmetry equation as it is the ultimate example of it. Cyber also has the added advantage/disadvantage of being hard/impossible to source (you can't watch the "cyber drones" slowly flying over from Iran). We urgently need to revisit our military priorities and spending in the US. In addition, we need to recognize that our historic advantages--natural resources, vastly superior technological lead, work ethic, education--may no longer exist and will require significant changes (cultural, political) to our society. This isn't just a military hardware issue.
I wonder how long it will be before we see functional Hunter-Seeker assassination drones? How will a politician defend against a few hundred of these insect sized drones with facial recognition?
A few hundered interns with fly swatters?
The huge lesson contained in the Georges Clemenceau quote...... "Generals always prepare to fight the last war." (although possibly apocryphal) springs to mind here.
This is usually meant as a criticism, but "preparing to fight the last war" is kind of something you have to do anyway even if you know the next war is probably going to end up being different in important ways. This is because there almost always someone else out there with the ability to fight the same kind of war as the last one, and you don't want them to try it. The only thing dumber than preparing for a war like the previous one is *not* preparing for it, having to do it anyway, and then *losing*.
I don't think you've thought clearly about the range of small machines. Aircraft carriers are after all largely merely a way of transporting smaller machines closer to targets than they can fly on their own. Why cannot Israel use small machines against Iran if they use larger machines to transport them closer before launching them?
For the last two years, I have helped transfer non-lethal technology including drones to forces in Ukraine.
Lethal autonomous weapons are currently deployed in Ukraine in order to hit targets despite the jamming that prevents human control of radio-operated vehicles.
The components that support autonomy (sensors, compute and power) take up space and currently impose a floor on how small these devices can go. But they are small enough, because small is relative, and the machines do not have to be as small as insects to nevertheless create a new way of war.
Their size is relative to the capacity of defense systems to detect them. It also matters in a larger sense, because small devices can be made cheaply, and their numbers can overwhelm defense systems. They can overwhelm those systems in two ways: 1) they can target a lot sites, more than the enemy has expensive systems to cover. 2) they can target a single site with more attack vehicles than the enemy has defense missiles.
Size and cost are generally correlated, although the relationship is not 1:1. That is, by shrinking the size of devices, more can be made on the same budget. This logic doesn't strictly hold for Western militaries, where the incentives are perverse, but it does in Ukraine, because Ukraine is fighting a poor man's war. By driving down cost, one country can produce equipment in quantities that overwhelm the opponent, and also set up an advantageous trade of low-cost equipment for high-cost. All things being equal, this is how a war of attrition can be won.
Unfortunately, the US defense procurement process is spec-driven and its minimum timeline from need to deployment is measured in years, by which time most equipment has been obviated. What the US must do if it is to defend itself or its allies is create fast feedback loops between technologists and battlefield users in hot wars, short-circuiting the DOD's spec-driven culture in order to iterate more quickly and make what is both necessary and asymmetric to the enemy's capabilities. This is currently not how defense tech is produced.
Redundancy is important- so too is repairability.
I think firms like Anduril and Palantir will help in the small warfare space long term.
Still need some serious investments in conventional naval warfare though- see here:
https://open.substack.com/pub/cdrsalamander/p/ill-take-10of-everything?utm_source=share&utm_medium=android&r=g90v
Apparently, Russian officers know command soldiers in Ukraine via radio while monitoring them with drones in real time. It’s basically Warcraft at this point.
YOU MUST CONSTRUCT ADDITIONAL PYLONS
(technically Starcraft, but close enough)
Depending on the theatre, if we consider the proposition at its scaled potential (i.e. mass swarms), then it becomes similar to an advance force ahead of occupying infantry. However, any infantry that occupies a territory will still be subject to mass counterattacks, likely from a stronger position when compared to the invading force. Strategically, you still need everything else - navy and air force for logistics and delivery of personnel, weapon systems, ammunition and resupply. To knock those systems out, you still need long range weapon strategic weapon systems.
Drones may offer a superior platform for localised dominance (eg over tank battalions), but not a significant strategic leap. Ukraine is seriously held back from using long range weapons, and we are still talking about small geographical distances. So they are forced into using drones.
In the Pacific, the most likely theatre against the CCP, you need strategic capability over vast distances. The next iteration is not likely drones, but hybrid space weapons or pure space deployed weapons. Countries use their non-defense industrial base advantages in total war - America's in the next war will be SpaceX, autonomy, software and AI. If you have cheap launch costs, you can have your war factories still position far away from the Pacific, launch vehicles into space, have them maneuver and deploy in a 30 minute window across the Pacific (example being the often discussed "Rods from God" concept [https://austinvernon.site/blog/starshipsuperweapon.html#:~:text=Rods%20from%20God&text=In%20the%201950s%2C%20a%20Boeing,feet%20underground%20to%20destroy%20bunkers.]
When you need to knock out factories, ports, refineries, rail, you need longer range deployment. Space is that channel. This also goes for directed energy weapons.
I can understand if you combine a Space based deployment with a cluster of drones that break apart from a single rod fired from space, and then swam locally to dominate a target. However, the hard part is the cheap reusable space deployment, not the drones.
So yes, drones will be valuable for smaller localised conflicts, but for the one that matters in the next thirty years - the Pacific - it will lend itself towards America's current advantages.
Slaughterbots (from 5 years ago): https://youtu.be/fPqmC16ewYg?si=nnp51_ju35KxfBjJ
Conventional wisdom right now seems to be that the key feature of drones is "attritability." Losing a drone is not very bad, since they're cheap and you don't lose the skilled pilot. So you can make more and riskier attacks using drones than using manned systems.
My son has been working in the area of very small and light fully autonomous drones. Only by being autonomous can you avoid electromagnet interference with communications.
He is looking at an initial commercial markets using them to defend farms against birds (he lives in Napa where birds like grapes making this a real issue). His final market is imagined at defending against small drone swarm attacks with smaller/cheaper drones.