Modern kamikaze drones are pretty devastating, routinely blowing up tanks and other primary weapons platforms in Ukraine. The infantrymen need to be lucky every time, the drone only needs to get through once. They're very cost-efficient.
In addition to No_one's fleets of bomb-droppers, spotters and kamikaze drones, I'll add self-propelled artillery pieces, minelaying artillery like Russia used to great effect, traditional high-altitude airpower, ballistic missiles, SAMs and some low-altitude drones carrying longer-range missiles like Hellfires (do we really need a whole Apache gunship anymore?) Maybe some tracked vehicles with LMGs to escort the vulnerable heavier vehicles against anything that slips through. These all seem fairly open to mechanization, at least more than legged infantry. There would be great dividends in fire control, coordinates of enemy targets would go from drone spotter to robotic artillery at machine speeds.
Jamming would be one countermeasure, yet jammers put out a great big 'here I am' signal. That's begging for an anti-radiation missile or artillery fire, just like how radar-guided SAMs need to watch out. Jamming would be a useful tool but not necessarily a hard counter. In Ukraine, drones can drop bombs from above the effective range of ECM mounted on vehicles, or fly in on a ballistic trajectory after control is lost.
Such robotic forces would probably flounder in urban warfare, where line of sight is low and there's plenty of cover available. Nothing stops them laying siege though. Relying on a central computing/data processing post is also a risk, I'm envisioning a huge truck or armoured train full of expensive compute. I suppose you could put a lot of air defence and guards nearby though.
Perhaps the biggest risk is cyberwarfare, losing command and control over one's robots.
Modern kamikaze drones are pretty devastating, routinely blowing up tanks and other primary weapons platforms in Ukraine. The infantrymen need to be lucky every time, the drone only needs to get through once. They're very cost-efficient.
You are describing an artillery shell.
So long as artillery shells can't by themselves hold positions and police an area infantry will remain the sole reason any other military implement exists.
There is this myth that has captured the imagination of American aligned armies after the cold war, that air superiority or any other kind of area denial of that sort means anything by itself.
The lesson of all of the recent American defeats is that if you don't have grunts patrolling some territory unmolested, you don't hold shit. You're just having an extended operation behind enemy lines.
Maybe robots will one day have the flexibility to act as grunts, but so long as they're still, in any mass producible form, glorified fire support, you still need infantrymen.
IGI-111
God was a dream of good government.
No_one 4mo ago
Neither do modern kamikaze drones that are built in significant scale like we're talking about here. The thermals alone would be more expensive than the whole thing.
Besides, reconnaissance is not a primary goal of warfare, and has long escaped the sole magisteria of grunts.
Satellites are rare, planes won't spot sneaky units and can be shot down anyway.
Drones with good thermals at say, company level are a complete change of the meta.
IGI-111
God was a dream of good government.
No_one 4mo ago
You already know what I'm going to say: the variants that have good optics are not kamikaze unless they're purpose built to destroy HVTs and are therefore not what we're talking about.
Smart artillery shells are a boutique implement, anything at that price is cool tech that's situationally useful but not meant for mass.
And my concern is that the nature of war is mostly defined by mass.
You're right to say that company level reconnaissance has been greatly improved and that it's a game changer. I would never deny (relatively) cheap optics isn't an important development.
Everyone is saying this is why Ukraine is an artillery war, because both sides have almost total information.
But again none of that is really what we're talking about when we talk about autonomous warfare. Because none of that changes the game in ways that create a novel equilibrium. We're just back to a state where defense has a large advantage and firepower beats maneuver. Which is certainly a change, but not one that has much to do with automation or robots in character.
Infantry absolutely are needed right now, for urban warfare, for reconnaissance, for fighting in forests. I've made posts about how the US needs infantry and recruits in the combat arms generally.
Where the US went wrong in its desert wars was that it went in without a clear plan of what to do after ejecting Saddam/Taliban from power. The political skills just weren't there to follow up victory on the ground. It's not that there was a shortage of infantry but that infantry weren't committed in a well-considered campaign with the right goals. They pumped a grossly corrupt and incompetent Afghan govt full of money, allied with the child rapist elements of the country - yet the goal was to turn Afghanistan into a liberal democracy. In Iraq they were catching and releasing terrorists (or torturing them in ways that made Arabs very angry). That doesn't fit with the war goal! The US is just really bad at a certain kind of martial imperialism, the British or Romans were much better at this kind of thing.
What I'm talking about is future conventional war, between serious countries that know what they're doing. A war with proper, realistic, military goals can be fought with firepower alone. In Desert Storm (a masterpiece in how wars should be fought), the firepower intensive elements of the Coalition won the victory, infantry were barely needed. There were some special forces that did useful work, some infantry engaged with their anti-tank weapons but it was mostly won by airpower, artillery and armour. There was a clear plan - thrash the Iraqi army on the battlefield till they leave Kuwait.
Holding positions - what does this mean? Sitting in a trench or hiding in a forest with rifles and ATGMs, ready to pop out and attack hostiles as they approach. Robots can do that in the future. Put an LMG on a little tracked vehicle or a bulked up Boston Dynamics bot, add some optics and that can hold positions fine, with the right software. Landmines can hold a position. You could leave some kamikaze drones in a forest on standby mode, they could potentially hold a position. Urban warfare makes things harder of course.
Policing is a different problem to these battlefield issues and the very name implies it's something for civilian police or military police, not infantry.
IGI-111
God was a dream of good government.
RandomRanger 4mo ago·Edited 4mo ago
Policing is a different problem to these battlefield issues and the very name implies it's something for civilian police or military police, not infantry.
I understand it's categorically different when we're talking about combat, but to this particular issue it's significant because it draws the boundary of things that robots can't do.
The reason robots are not going to replace the infantryman at the margins is because they, in their current incarnations, are incapable of improvisation or dealing with general problems. They are indeed just a more sophisticated version of a mine.
And no, a mine can't hold a position. It can slow the enemy, it can increase attrition, it can funnel the enemy where you want him or free up your forces to be used elsewhere. It can do a lot of useful things, but it can't hold a position.
Loitering munitions can probably even make what used to be static defense a lot more mobile, but mobile or not it's still a castle/trap. You need men behind those supposedly automated defenses or the enemy is going to exploit it because robots can't adapt on the fly and can be jammed or sidestepped. Even drones can do so a lot less than actual boots on the ground.
Look at Hamas' low tech tour de force which was all against state of the art automated defenses. It's emblematic of what you do to deal with that kind of thing: you find some exploit, sit on it for a while, and nullify the defenses all at once when the enemy expects they're solid.
Put an LMG on a little tracked vehicle or a bulked up Boston Dynamics bot, add some optics and that can hold positions fine, with the right software
I'm well aware of what can be done, friend of mine actually writes that exact kind of software. And it's not magic, it's just a much more annoying claymore at the end of the day. Bots don't dig foxholes or deal with complex terrain very well. Their best use is in in freeing up hands to do other things.
I'm skeptical you could effectively use automated turrets on the battlefield in a way that wouldn't eventually be nullified because I think the environment is too chaotic.
What do you mean by holding a position? Entrenched infantry slows down the enemy, inflicts losses, funnels them where you want them. The primary difference is that infantry is mobile, yet drones can also be mobile. Engineering vehicles can dig earthworks suitable for tanks. A mini-tank could presumably go hull-down, though it seems most threats come from above these days.
What good are entrenched infantry going to be when a swarm of kamikaze drones fly down into their trenches at 3 in the morning?
Assuming we've got them all linked up to an AGI performing the role of brigade command, robotic forces could adapt and execute plans, counter exploits. Wearing weird camouflage patterns probably wouldn't work on a decent AGI - if all else fails they could use thermal imaging. Palantir has already made a test LLM for quickly organizing strike missions, I see no reason why machines couldn't execute tactics. The speed and coordination of an unmanned force would be extremely impressive. Commanders are deluged in information from all the sensors in modern warfare, machines are best at managing a tsunami of data and providing quick answers. The gains in cost-efficiency and speed will probably outweigh the loss in human flexibility for most environments. I'll admit that infantry will be better in urban environments for a long time to come. But everyone seems to agree that urban environments are hellish to fight in, it might be easier to encircle and siege them out.
Jamming isn't a foolproof answer. The transmitter will be lit up for artillery or missile fire. Some weapons could be designed to go into an autonomous mode if they lose connection to command and control.
Look at Hamas' low tech tour de force which was all against state of the art automated defenses
Au contraire, the problem was that the human guards were understrength and unprepared due to the holiday. Arrogance and complacency was their core problem. They had a bunch of sensors but relied on a response force of human soldiers that simply wasn't there. There weren't any landmines beneath the fence (or certainly not enough to impede Hamas planting explosives there). There wasn't a response force ready to go, they planned assuming a 24-hour warning time to deploy.
For at least three months prior to October 7, the soldiers recalled reporting information on Hamas operatives conducting training sessions multiple times a day, digging holes and placing explosives along the border.
However, when presenting the evidence to their senior officers, they were ignored, and the information was not passed further up the chain of command.
While there were three infantry battalions and one tank battalion positioned along Gaza’s border, stated the report, a senior military officer estimated that perhaps half of the 1,500 soldiers were away.
Now maybe an AI-based combat system would also have failed here, yet the humans certainly didn't cover themselves in glory. Sensors are useless if the humans don't listen to them and aren't prepared to act. They missed the warnings, they fielded understrength forces, they waited too long to deploy and were uncoordinated in their initial attacks (since divisional HQ was directly attacked). Hamas also made good use of drones to defeat turrets and armour, which doesn't necessarily counter my argument.
IGI-111
God was a dream of good government.
RandomRanger 4mo ago·Edited 4mo ago
What do you mean by holding a position?
Continuously effect control in the tactical sense "maintain physical influence over a specified area to prevent its use by enemy or to create conditions necessary for successful friendly operations" as the Americans put it.
But we're about to get in pedantic debates about things being tools of control instead of effecting control in and of themselves, because the real disagreement we are having, I think, is as to considering automated systems as agents. I do not think it is wise to do so.
the problem was that the human guards were understrength and unprepared due to the holiday
I think this is the key to what I'm saying. The fact that the human element failed is evidence that it is still and will probably remain crucial to effective operations even in a world with drones.
If drones were effective on their own and could replace the infantryman, the lack of intelligence would not have mattered as the automated defenses could have adapted to the breach. But machines are not generally intelligent so they can't do that, you need some human to figure it out before he gets killed, and the fewer humans you have right next to the problem the harder it is to figure things out on the fly.
I do not mean to imply that automated systems won't play a big part as a force multiplier in wars current and future. I only mean to assert that you're not going to be able to remove the humans from direct battlefield involvement without compromising capability to an unreasonable degree.
This is why I've long prognosticated that flesh and blood pilots would continue to exist in significant numbers even as drones become more sophisticated. There's just too many situations where looking at the objective or the enemy with your eyes and making a decision right there and then beats interpreting a camera feed, even with how sophisticated modern sensors are.
And ground combat is a lot more chaotic and difficult to parse so I intuit it's even less likely.
Another factor we haven't really discussed is how fragile all this equipment is. For all the fictional cachet robots get for being indestructible machines of death, they are surprisingly brittle and needy. If we had to bet on who survives continuous shelling for the longest, I'm not sure I'm putting my money on the high tech stuff. Even if it's a more morally acceptable loss, obviously.
This website is a place for people who want to move past shady thinking and test their ideas in a
court of people who don't all share the same biases. Our goal is to
optimize for light, not heat; this is a group effort, and all commentators are asked to do their part.
The weekly Culture War threads host the most
controversial topics and are the most visible aspect of The Motte. However, many other topics are
appropriate here. We encourage people to post anything related to science, politics, or philosophy;
if in doubt, post!
Check out The Vault for an archive of old quality posts.
You are encouraged to crosspost these elsewhere.
Why are you called The Motte?
A motte is a stone keep on a raised earthwork common in early medieval fortifications. More pertinently,
it's an element in a rhetorical move called a "Motte-and-Bailey",
originally identified by
philosopher Nicholas Shackel. It describes the tendency in discourse for people to move from a controversial
but high value claim to a defensible but less exciting one upon any resistance to the former. He likens
this to the medieval fortification, where a desirable land (the bailey) is abandoned when in danger for
the more easily defended motte. In Shackel's words, "The Motte represents the defensible but undesired
propositions to which one retreats when hard pressed."
On The Motte, always attempt to remain inside your defensible territory, even if you are not being pressed.
New post guidelines
If you're posting something that isn't related to the culture war, we encourage you to post a thread for it.
A submission statement is highly appreciated, but isn't necessary for text posts or links to largely-text posts
such as blogs or news articles; if we're unsure of the value of your post, we might remove it until you add a
submission statement. A submission statement is required for non-text sources (videos, podcasts, images).
Culture war posts go in the culture war thread; all links must either include a submission statement or
significant commentary. Bare links without those will be removed.
Jump in the discussion.
No email address required.
Notes -
Modern kamikaze drones are pretty devastating, routinely blowing up tanks and other primary weapons platforms in Ukraine. The infantrymen need to be lucky every time, the drone only needs to get through once. They're very cost-efficient.
In addition to No_one's fleets of bomb-droppers, spotters and kamikaze drones, I'll add self-propelled artillery pieces, minelaying artillery like Russia used to great effect, traditional high-altitude airpower, ballistic missiles, SAMs and some low-altitude drones carrying longer-range missiles like Hellfires (do we really need a whole Apache gunship anymore?) Maybe some tracked vehicles with LMGs to escort the vulnerable heavier vehicles against anything that slips through. These all seem fairly open to mechanization, at least more than legged infantry. There would be great dividends in fire control, coordinates of enemy targets would go from drone spotter to robotic artillery at machine speeds.
Jamming would be one countermeasure, yet jammers put out a great big 'here I am' signal. That's begging for an anti-radiation missile or artillery fire, just like how radar-guided SAMs need to watch out. Jamming would be a useful tool but not necessarily a hard counter. In Ukraine, drones can drop bombs from above the effective range of ECM mounted on vehicles, or fly in on a ballistic trajectory after control is lost.
Such robotic forces would probably flounder in urban warfare, where line of sight is low and there's plenty of cover available. Nothing stops them laying siege though. Relying on a central computing/data processing post is also a risk, I'm envisioning a huge truck or armoured train full of expensive compute. I suppose you could put a lot of air defence and guards nearby though.
Perhaps the biggest risk is cyberwarfare, losing command and control over one's robots.
You are describing an artillery shell.
So long as artillery shells can't by themselves hold positions and police an area infantry will remain the sole reason any other military implement exists.
There is this myth that has captured the imagination of American aligned armies after the cold war, that air superiority or any other kind of area denial of that sort means anything by itself.
The lesson of all of the recent American defeats is that if you don't have grunts patrolling some territory unmolested, you don't hold shit. You're just having an extended operation behind enemy lines.
Maybe robots will one day have the flexibility to act as grunts, but so long as they're still, in any mass producible form, glorified fire support, you still need infantrymen.
Artillery shells don't have thermals and can't report what's going on under them.
Neither do modern kamikaze drones that are built in significant scale like we're talking about here. The thermals alone would be more expensive than the whole thing.
Besides, reconnaissance is not a primary goal of warfare, and has long escaped the sole magisteria of grunts.
They're already in use.
The recon drone has good thermals, costs less than smart artillery shells, the fpvs have shitty. (looks like 320x240) or less.
Lancet UAV has a thermals variant.
Satellites are rare, planes won't spot sneaky units and can be shot down anyway. Drones with good thermals at say, company level are a complete change of the meta.
You already know what I'm going to say: the variants that have good optics are not kamikaze unless they're purpose built to destroy HVTs and are therefore not what we're talking about. Smart artillery shells are a boutique implement, anything at that price is cool tech that's situationally useful but not meant for mass. And my concern is that the nature of war is mostly defined by mass.
You're right to say that company level reconnaissance has been greatly improved and that it's a game changer. I would never deny (relatively) cheap optics isn't an important development. Everyone is saying this is why Ukraine is an artillery war, because both sides have almost total information.
But again none of that is really what we're talking about when we talk about autonomous warfare. Because none of that changes the game in ways that create a novel equilibrium. We're just back to a state where defense has a large advantage and firepower beats maneuver. Which is certainly a change, but not one that has much to do with automation or robots in character.
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
Infantry absolutely are needed right now, for urban warfare, for reconnaissance, for fighting in forests. I've made posts about how the US needs infantry and recruits in the combat arms generally.
Where the US went wrong in its desert wars was that it went in without a clear plan of what to do after ejecting Saddam/Taliban from power. The political skills just weren't there to follow up victory on the ground. It's not that there was a shortage of infantry but that infantry weren't committed in a well-considered campaign with the right goals. They pumped a grossly corrupt and incompetent Afghan govt full of money, allied with the child rapist elements of the country - yet the goal was to turn Afghanistan into a liberal democracy. In Iraq they were catching and releasing terrorists (or torturing them in ways that made Arabs very angry). That doesn't fit with the war goal! The US is just really bad at a certain kind of martial imperialism, the British or Romans were much better at this kind of thing.
What I'm talking about is future conventional war, between serious countries that know what they're doing. A war with proper, realistic, military goals can be fought with firepower alone. In Desert Storm (a masterpiece in how wars should be fought), the firepower intensive elements of the Coalition won the victory, infantry were barely needed. There were some special forces that did useful work, some infantry engaged with their anti-tank weapons but it was mostly won by airpower, artillery and armour. There was a clear plan - thrash the Iraqi army on the battlefield till they leave Kuwait.
Holding positions - what does this mean? Sitting in a trench or hiding in a forest with rifles and ATGMs, ready to pop out and attack hostiles as they approach. Robots can do that in the future. Put an LMG on a little tracked vehicle or a bulked up Boston Dynamics bot, add some optics and that can hold positions fine, with the right software. Landmines can hold a position. You could leave some kamikaze drones in a forest on standby mode, they could potentially hold a position. Urban warfare makes things harder of course.
Policing is a different problem to these battlefield issues and the very name implies it's something for civilian police or military police, not infantry.
I understand it's categorically different when we're talking about combat, but to this particular issue it's significant because it draws the boundary of things that robots can't do.
The reason robots are not going to replace the infantryman at the margins is because they, in their current incarnations, are incapable of improvisation or dealing with general problems. They are indeed just a more sophisticated version of a mine.
And no, a mine can't hold a position. It can slow the enemy, it can increase attrition, it can funnel the enemy where you want him or free up your forces to be used elsewhere. It can do a lot of useful things, but it can't hold a position.
Loitering munitions can probably even make what used to be static defense a lot more mobile, but mobile or not it's still a castle/trap. You need men behind those supposedly automated defenses or the enemy is going to exploit it because robots can't adapt on the fly and can be jammed or sidestepped. Even drones can do so a lot less than actual boots on the ground.
Look at Hamas' low tech tour de force which was all against state of the art automated defenses. It's emblematic of what you do to deal with that kind of thing: you find some exploit, sit on it for a while, and nullify the defenses all at once when the enemy expects they're solid.
I'm well aware of what can be done, friend of mine actually writes that exact kind of software. And it's not magic, it's just a much more annoying claymore at the end of the day. Bots don't dig foxholes or deal with complex terrain very well. Their best use is in in freeing up hands to do other things. I'm skeptical you could effectively use automated turrets on the battlefield in a way that wouldn't eventually be nullified because I think the environment is too chaotic.
What do you mean by holding a position? Entrenched infantry slows down the enemy, inflicts losses, funnels them where you want them. The primary difference is that infantry is mobile, yet drones can also be mobile. Engineering vehicles can dig earthworks suitable for tanks. A mini-tank could presumably go hull-down, though it seems most threats come from above these days.
What good are entrenched infantry going to be when a swarm of kamikaze drones fly down into their trenches at 3 in the morning?
Assuming we've got them all linked up to an AGI performing the role of brigade command, robotic forces could adapt and execute plans, counter exploits. Wearing weird camouflage patterns probably wouldn't work on a decent AGI - if all else fails they could use thermal imaging. Palantir has already made a test LLM for quickly organizing strike missions, I see no reason why machines couldn't execute tactics. The speed and coordination of an unmanned force would be extremely impressive. Commanders are deluged in information from all the sensors in modern warfare, machines are best at managing a tsunami of data and providing quick answers. The gains in cost-efficiency and speed will probably outweigh the loss in human flexibility for most environments. I'll admit that infantry will be better in urban environments for a long time to come. But everyone seems to agree that urban environments are hellish to fight in, it might be easier to encircle and siege them out.
Jamming isn't a foolproof answer. The transmitter will be lit up for artillery or missile fire. Some weapons could be designed to go into an autonomous mode if they lose connection to command and control.
Au contraire, the problem was that the human guards were understrength and unprepared due to the holiday. Arrogance and complacency was their core problem. They had a bunch of sensors but relied on a response force of human soldiers that simply wasn't there. There weren't any landmines beneath the fence (or certainly not enough to impede Hamas planting explosives there). There wasn't a response force ready to go, they planned assuming a 24-hour warning time to deploy.
SIGINT apparently wasn't working on holidays: https://www.timesofisrael.com/top-israeli-intel-unit-wasnt-operational-on-october-7-due-to-personnel-decision/
Now maybe an AI-based combat system would also have failed here, yet the humans certainly didn't cover themselves in glory. Sensors are useless if the humans don't listen to them and aren't prepared to act. They missed the warnings, they fielded understrength forces, they waited too long to deploy and were uncoordinated in their initial attacks (since divisional HQ was directly attacked). Hamas also made good use of drones to defeat turrets and armour, which doesn't necessarily counter my argument.
Continuously effect control in the tactical sense "maintain physical influence over a specified area to prevent its use by enemy or to create conditions necessary for successful friendly operations" as the Americans put it.
But we're about to get in pedantic debates about things being tools of control instead of effecting control in and of themselves, because the real disagreement we are having, I think, is as to considering automated systems as agents. I do not think it is wise to do so.
I think this is the key to what I'm saying. The fact that the human element failed is evidence that it is still and will probably remain crucial to effective operations even in a world with drones.
If drones were effective on their own and could replace the infantryman, the lack of intelligence would not have mattered as the automated defenses could have adapted to the breach. But machines are not generally intelligent so they can't do that, you need some human to figure it out before he gets killed, and the fewer humans you have right next to the problem the harder it is to figure things out on the fly.
I do not mean to imply that automated systems won't play a big part as a force multiplier in wars current and future. I only mean to assert that you're not going to be able to remove the humans from direct battlefield involvement without compromising capability to an unreasonable degree.
This is why I've long prognosticated that flesh and blood pilots would continue to exist in significant numbers even as drones become more sophisticated. There's just too many situations where looking at the objective or the enemy with your eyes and making a decision right there and then beats interpreting a camera feed, even with how sophisticated modern sensors are.
And ground combat is a lot more chaotic and difficult to parse so I intuit it's even less likely.
Another factor we haven't really discussed is how fragile all this equipment is. For all the fictional cachet robots get for being indestructible machines of death, they are surprisingly brittle and needy. If we had to bet on who survives continuous shelling for the longest, I'm not sure I'm putting my money on the high tech stuff. Even if it's a more morally acceptable loss, obviously.
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link