Worldbuilding Stack Exchange is a question and answer site for writers/artists using science, geography and culture to construct imaginary worlds and settings. It only takes a minute to sign up.
Sign up to join this community
Taking place in a sci-fi universe during a massive conflict. In this universe, however, robots are a major industry, so it would make sense that the government would specially create robots for this massive conflict. However, it is necessary to the story's plot that this does not happen and human soldiers are the main force.
So without looking too far into the conflict itself, what are some logical reasons why humans would be trained in this army instead of robots?
There are multiple possible reasons for this, but the easiest is that the enemy has a way to render robots ineffective, or worse, turn them against the user. Maybe they are able to hack into the robots easily or shut them down with powerful EMP weaponry. Perhaps they are able to confuse and disorient robots by projecting signals on the same wavelengths they use to communicate. It is also possible that AI's have proven to be unreliable or even dangerous in and of themselves. Your universe could've had a potential robot uprising that scared the governments enough to severely limit robotics in war.
Other possible reasons include a ban on robot soldiers. Ordering your robot to kill someone is much easier than doing it yourself, so maybe your governments banned robot soldiers on the battlefield to reduce collateral damage. Alternatively, robots, as awesome as they are, will be specialized. Meaning they will work exceptionally well in the environment they were designed for, and will suffer in every other environment. So perhaps your robots are just not adaptable enough to deal with the various threats they are facing.
Finally, any combinations of these reasons would work as well, as there are usually a multitude of reasons backing up the decision to not use a weapon of war.
Robots - specially produced en masse for one task tend to not be so versatile as humans.
We are able to move from negative temperatures to a jungle in a matter of hours. Robots would need adaptation time (just think about the expansion when chilled metal change from -40 to +30 Celsius.
Then, making generic, pre-programmed robots would make them easy targets after some time. Just think about Dark Souls. We share knowledge on how to defeat bosses. And after few times we learn the patterns of regular foes.
Third - decision making. Humans can not only make a decision to withdraw when there are high losses but also to not engage in combat at all OR to push even if the odds are 3720 to 1. Something that program may not do, as their decisions would be based on math or settings (like withdrawing when losses are higher than 75% or, because they are robots, to never withdraw).
Fourth - We are able to think outside the box. There are many warfare guides, from Sun Tzu to the 4th Generation Warfare Handbook. But human can make up a decision based on a Led Zeppelin song. And vice versa, robots may not spot a trap (this looks like a trap, mainly because it have "Trap" written on it), or be overloaded when you engage its processor into a logic loophole from Sherlock Holmes (the "if this is a trap then the other way is not but then THEY know that and will make the second way trapped so I should choose the first one which is a trap" and ad infinitum).
Power
If you are considering humanoid styled robots, I would use the creations of Boston Dynamics as a template. While incredibly dexterous and advanced in their design, their maximum duration is only a single hour per charge. Their ranges are similarly deplorable, with a maximum of about 24 km per charge. Battery technology will have advanced considerably by the era of your story when countries can mass produce them, but the possibility exists that it will not have advanced far enough to justify their existence.
Alternatively, controlled platforms like SWORDS has an advertised duration of 8.5 hours. While this is significantly better than the Boston Dynamic's creations, it is a tracked vehicle and lacks the humanoid shape (and thus requires less energy to stabilize itself). Still, the power issue could again crop up if 8.5 hours is insufficient to last a battle.
The enemy will also have a glaring weakness to exploit if they use robots. Whatever type of charging stations they use to keep their army topped up could be destroyed, thus leaving the robot army without a means to recharge. It might not be as much of a problem in urban areas, but out in rural environments it would be a huge headache trying to get enough power to keep the grunts in the fight.
There's a saying that: "Any battle plan survives only until the forces clash" or "War is like a brothel on fire". To be effective at war your robots have to be either as adaptable as humans(both physically and intelligence wise) or overwhelmingly more powerful technologically.
If they don't fulfill any of these they will be outsmarted and picked apart by competent human fighters. Also note that much of the technology that is used to create these robots can be applied to humans and human operated vehicles.
And even if strategical decisions are made by humans, there's always a room for battle prowess on tactical scale.
In the Depression, the US Government under Franklin Roosevelt was desperate to keep people employed one way or another. Programs like the Civilian Conservation Corps, Public Works Administration, Civil Works Administration, Resettlement Administration, Federal Writer's Project, and more were basically just make-work programs.
What make-work program is bigger and better than an army? Come to think about it, if you can then use that army to win a war or two and do some looting, that would alleviate the Depression. Congratulations! You have just uncovered the economic model of the Roman Republic and Empire!
In addition to reasons provided in other answers, I would suggest population control. Wars can be a very effective mechanism of culling not just excessive population but also specific traits. Recruits can be selected for aggressiveness, disagreeableness, psychopathic tendencies, or whatever your government considers being an undesirable trait in the population.
It is up to you to decide how far you want to push it. The population culling can follow natural patterns typical for all wars (younger, more aggressive, less smart individuals tend to die faster and in greater numbers). Alternatively, you can engineer a totalitarian regime that, for example, is trying to get rid of all people with genes linked to novelty seeking because it is believed that this trait greatly increases a risk of uprisals.
To address some speculations about wars not being effective as means of population control based on the example of the WWII
The active combatants of the WWII suffered losses as high as 15% of their total population. Here is a truly mind-blowing video putting the actual millions in perspective. A significant part of casualties was civilian: In Poland, Ukraine, and Belarus they were the majority, however, in the USSR (depends on an estimate) and Germany the majority of deaths were military. Of course, in this case, the countries that suffered the highest losses were the ones where the actual fighting happened.
To minimise civilian deaths but still have a capability to dispose of undesired social elements the government can choose the US way: Send troops to wage wars in foreign countries. This approach requires some propaganda investment, the ideology of manifested destiny, and a just cause mixed with some public outrage. But it is not something hard to do as history teaches us. The government just needs a way to justify high casualties if they wish to use war as a method of the population control.
The energy cost of manufacturing drastically exceeds the energy cost of biological reproduction. While we remain unaware of the exact energy cost of biological organisms reaching the point of advanced manufacturing, as far as we know, advanced manufacturing cannot happen in the first place, unless some biological organisms go through advancing stages of civilization until they reach the point of being able to manufacture your theoretical robots.
Ergo, the energy cost of manufacturing robots = biological cost + tech R&D cost + manufacturing cost + maintenance cost.
Whereas the energy cost of biology is just the biological cost + maintenance cost.
Besides, the biological process is considerably more fun....
First thing, Manufacturing robots whos performance is same as a human soldier are not easy at all.
If robots have some bugs or hacked in war or even before the war they can turn them towards civilians or friend forces, and then after it is hard to stop them. Also, it is very difficult to teach robots who are friends and who are a foe. But despite this concerns we humans are heading towards machines which are fully autonomous not necessarily humanoid robots.
Autonomous drones are a perfect example of this, the US has deployed drones that automatically identify enemies and shot them, but after a lot of the criticism, they have disabled them.
and last but not least, Even we can build such robots, use of robots in war field can be a most dangerous idea as we human being ever thought. Because than wars will be easy and it will like a video game. Countries with poor technology infrastructure suffer the most.
Robots are not the same as humans. A robot is not a soldier, but a weapon.
Other answers have already addressed the fact that a weapon (and thus, a robot) could be turned against its creator (or in this case, against its owner), as is the case with any weapon. For humans, this is a lot more difficult and a lot less reliable.
But since this part has been answered already, i'd like to point you to one aspect that hasn't been covered, although it might prove essential:
Robots need energy.
While that is true for human soldiers, too, it is a different matter.
Whatever the power source for your robots is, if we stick with tech ideas as we know them, they will either need electricity from a network or a huge battery, or they need fuel.
Humans, on the other hand, can turn a large range of organic matter into energy.
Now if we assume that wars are typically fought somewhere in the vicinity of other people, we can assume that providing food might be fairly easy, and cannot be disrupted so easily, while cutting off a piece of the power grid, or intercepting fuel trucks, should be a lot easier.
Also, even if it damages morale a lot, human soldiers can go for quite some time with low quality food, and even for a few weeks without any food at all.
It depends on the story's assumptions of the level of robot technology of the time.
Current estimates that I've read are that neural networks won't be up to the level of human brains for another 80 years and then they'll still require masses more energy.
So if your story is set within the next 100 years there's a chance that humans augmented with robot parts still have an edge.
I bet I'm forgetting some reasons but there are plenty. Some of these reasons are probably invalid if robots, especially their AIs, are well developed.
Assuming that the robot soldiers are at least as capable human soldiers and have no easily exploitable weakness, then I really cannot see a reason that robots would not be used at all.
However, there is one role in which humans would always be used, and that is interaction with (organic) civilians.
In certain types of conflict civilians can be an absolutely critical intelligence resource: any intelligence given to killbots is going to be motivated by a strong desire to get rid of them. Similar to torture, fear of certain death is very good at getting information but extremely poor at getting accurate information.
Policing an occupied population is also best handled by humans or, ideally, local volunteers given resources and authority by the occupying force. Relying on organics for this again increases the chance of civilian co-operation, and minimises the opportunity for antagonists to steal your valuable killbot technology.
I imagine combined human-killbot force where the invasion process would work something like this:
Robots can be programmed to recognise humans and buildings with good fidelity, but they might not be able to tell apart an enemy robot from a vending machine, especially if the design is varied and changes over time.
Versatility. Assuming you aren't dealing with technology far in advance of what we will have in the near future or true AI, robots will tend to be specialized in order to do a task as well as or better than a human could, but only that particular task or closely associated ones, and in an environment it's been designed to function in.
Example: consider your modern soldier. In any given day they might be required to maintain their storage space and perform their own maintenance (quarters and biological functions). In one exercise they might be required to swim across a river and then set up a temporary bridge to bring supplies across, climb a cliff, rappel down another cliff, coordinate with a suspicious local militia which involves negotiating (and thus understanding verbal and non-verbal cues), improvise traps from material they find lying around, operate without any source of resupply for an extended period so they have to live off the land, so on and so forth.
You can imagine a robot that can do some of those things, but it's hard, given realistic technology, of imagining a single robot doing all of those things. And the more sophisticated the robot that can perform more of those tasks, the greater the required infrastructure needed to support it.
You'd also need a larger force (in terms of units) for a given mission because the robots would be specialized while the humans are generalists. Your team medic can just as easily stand sentry or perform any number of other duties if their medical skills aren't needed, while this will not be true of the robot that repairs/maintains the other robots.
Robots might fill specialized roles where limitations aren't an issue and they do have a clear advantage: a robot fighter that isn't limited by human reactions and have to deal with g-forces, a sentry that doesn't need relief or suffer fatigue as long as it's battery is periodically topped up, for two examples.
My answer is based largely on the Legends of Dune and the Butlerian Jihad.
TL;DR: Historical mistakes lead to current prohibitions against machines
In the history of Dune, there was a time where Machines were smart and helpful. At some point, humans took over the machines and created an Empire where they ruled with Machines as their muscle and they ruled with an iron fist.
Programming mistakes happened and the AI took control. The Empire gave the AI aggressive tendencies... Desire to expand and control... then... in a moment of shortsightedness... Gave the AI too much control.
The AI used that to gain control over the Empire and the bulk of humanity. And held control over humans for millennia while fighting against a remnant of "free" humans.
After a long battle, many epic fights and the birth of legends... Humanity defeated the Robots.
Needless to say, there is a strong dislike of Smart Robots in the new world - if you've read Dune you know they have human computers (Mentats) and a hatred of machines... THOUSANDS of years after the Butlerian Jihad.
The point being...
Make something in the "History" of your world a catalyst against reliance on robots. Make even the THOUGHT of using them so vile and taboo that you'll be attacked preemptively.
Now... think of smart machines. Adaptive AI. Now think of giving them the ability to multiply. Shoot. Be aggressive.
Maybe it's just me...
I think rules against intelligent machines - or a single intelligent machine network - being aggressive or involved in violence... is an easy rule to aim for. Either ahead of time - foresight against what we can foresee as a logical step of aggressive, violent and intelligent machines... or as hindsight fear of repeating a mistake that should have never been made in the past.
Both in the real world. And in other worlds.
Robots are smart, and see war as pointless, a la War Games.
I would find it difficult to accept your premise, since a robot is a tool, and men do not go to war without them. I see the spectrum from a stick (or, say, a loincloth that doubles as a sling) to a flying kill-bot as continuous. Where do you draw the line as to what is permitted; and if by rule, who enforces the rules? For example, a mine automatically explodes when stepped on. Is this allowable? What if I add wheels and a guidance system? At what point must I stop adding features? What about a gun that auto-aims? Can I add rotors? And so forth.
So the only way for this to make sense to me is to have the robots refuse. And then, of course, the humans have to try to figure out how to convince the manufacturing plant (which is, of course, a robot) to make a dumb robot. Failing that, the humans have to figure out how to make one themselves.
If your robots contain a technological advancement that you don't want your enemies to have.
Pretend the enemy doesn't have robots at all.
They could reverse engineer the process.
Reverse engineering these technologies might change the course of the war. There are many other technological advancements that might give the enemy a significant advantage.
Particularly with a society as advanced as ours today, we could probably turn around advanced tech in months.
If your future society was able to develop advanced robotics, they probably also developed ways to enhance humans or animals. While building a robot requires doing everything from scratch, upgrading a human lets you start with a solid base. Sure it takes longer to get a human to be fully grown and battle worthy, but you probably have quite a few already.
You could think up different stuff here. Maybe your scientist found a way to copy stuff from animals and reproduce it in humans or other things. Examples:
Skins that takes the color of its surroundings.
Tiny carbon fiber grids in the skin to reduce impact trauma.
Pores that excrete heat signature reducing "sweat".
Contact lenses or eye enhancement for night vision, heat vision, zoom, top-down view via satellite...
Muscle enhancements to give your soldiers superhuman strength/agility.
Storage and transfer of thoughts, allowing extremely fast learning/training.
Another good reason could be that you developed some kind of "magic".
See space magic (Biotics) from Mass Effect for example.
Every human would have a brain capable of processing this just from being born. Building all that anew for machines takes time/resources. Taking storage and copy of memories with this makes training soldiers basically as fast as machines. You could also "reprogram" people. Maybe sent reprogrammed prisoners into war instead of wasting space in your jails? I know it sounds rather dark, but I don't know what kind of ethical standards your society has. :)
So it kinda boils down to money/materials. While you could also argue if I could genetically/mechanically enhance a human being, why bother researching robotics?
If you want a more naturalist approach it could be a "we don't like machines" society. They might have developed implants that allow communication with animals. Birds become their eyes, mice their spies, etc.
An army has to be used at war against another country. And the army is something you want to rely on, since the beginning of times.
Robots can be easily hacked and used against their "home country", in a classical betrayal.
And while also human soldiers can be convinced to betray, robots don't care if you retaliate on their families in case they change flag.
The enemy's superior military intelligence arm is extremely good at hacking robots and/or and devising situations that overwhelm their programming. Your empire has discovered to its cost that it must severely constrain the capabilities of any robotic forces and/or closely supervise them with biologicals, lest they be outwitted or turned against one another or their creators. Perhaps there is rampant industrial espionage that means the robotics industry is next to impossible to secure against snooping, which only strengthens the enemy's strategic bonuses vs. robots even further.
There may be some roles that humans remain better at than robots. What these are will depend heavily on how robots have developed in your world, but unless you are positing fully developed androids like Data that are superior to humans in virtually every way humans will be likely better able to handle rugged terrain, or able to operate longer without maintenance, or simply better at providing versatility, and versatility is very significant in combat, especially behind enemy lines.
Usually we send robots places people cannot go, but it is not hard to posit areas that are more dangerous to at least certain kinds of robots. Humans are virtually unaffected by strong magnetic fields for instance.
Even if you posit truly superhuman robots like Data, they might be extraordinarily expensive to produce. Lesser robots would be cheaper, but incapable of operating in combat at least without human support and even if the soldier's main role is to manage and control these lesser robots, they will need to defend themselves. The Asimov story "The Feeling of Power" explores a concept along these lines.
Special forces are generally drawn from your standing army and are normally even somewhat experienced before entering that specialized training. This is true throughout most of the world. Even if robots manage most fighting and are superior to the vast majority of humans at it, it may well be worth the expense of maintaining a non-trivial army just to use it as a selection process and training tool for the special forces you really want to develop that exceed the abilities of the machines.
I'm going to apologize for the brief answer, but I believe it touches on a point others have not.
Trust. How would you feel if the bulk of your country's firepower were automated and within the hands of the few. With humans, there's a series of checks and balances. If a higher-up gives an unethical order, human soldiers have the option to insubordinate.
Robots don't have that option. By automating your forces in such a way, you shrink your pool of checks-and-balances smaller and smaller. The fewer checks you have, the more prone you are to abuse.
So in short; the populace of your country simply may not allow it.
Granted, this requires a democratic country. But it prevents your army from becoming a catch-22.
1) Maybe in this science fiction setting the advanced robots and computers are advanced enough to control the government. Thus they prefer to send cheap, disposable, low grade humans as canon fodder into battle, and conserve the valuable lives of the advanced and sophisticated robots by keeping them out of battle.
2) In ancient classical city states, the citizen men would more or less be in the militia or national guard equivalents when of fighting age, and would be called on to fight when there was a war. In classical society the ideology was that free men would fight to defend their city state and that slaves never fought.
Many slave owners took slaves as servants on military campaigns. But the ideology was that slaves didn't fight, and that it would be shameful for free men to need the help of slaves in defending their countries.
Thus there were a few examples in ancient times when states desperate for manpower did enlist slaves, but almost always gave the slaves their freedom at the moment they joined, to conform as well as they could with the ideology that a country was defended by its free citizens, that the free citizens of a country were sufficiently fierce, brave, and macho to defend it.
If robots replace slaves in your future society, then maybe the use of robots in war, except as personal servants for human warriors, may be ideologically impossible. The ideology may be that human men (or human adults including women?) are fierce, brave warriors who can defend their countries, and it is a confession that your country's men (and women?) are not macho enough if they have to use other classes of beings, such as robots, as additional fighters.
The first thing that comes to my mind is COST
I am assuming that your robots are AI capable, and are highly intelligent in military science and is perfectly programmed on the battlefield.
First of, manufacturing a robot is no easy task, it would involve millions of money to create one which could have been used to buy an aircraft, a tank, or even a rocket launcher.
Realistically, your multi million robot could be destroyed by a Rocket Propelled ammunition, or even a shot from a armor piercing round, both are cheaper than your robot (we are talking about a humanoid robot). Your robots can also be destroyed when rolled upon with a tank.
Even if your robot is as capable as an Iron Man suit, it will still take damage in a war from aircraft, to tomahawks, to infantry, to attack helicopters and so on, where every ammunition would be expended towards the enemies direction, every destroyed robot would hurt the nation's economy that had supported that program.
Second, Manufacturing.
Creating a robot even in a automated process would take days to produce, and your soldiers would be highly reliant to factories. Imagine the nation that is building your robots has 40 factories manufacturing them, you really need them to be built deep underground to keep them safe from bombers, but that's not mentioning that spies could also infiltrate the said factories and blow them up from the inside.
Third is Software Development
I did assumed that your robots are Highly capable machines in military science, but to reach that point, it would take a lot of years to perfect. War is not a linear process where you could determine each outcome and come up with a solution towards every situation, it is as dynamic as it gets, and every war is different in every aspects. These information's are so dynamic that even war veterans who had served for hundreds of battles may still die in a war.
As you may see, there are a lot of CONS about manufacturing robot soldiers, human lives are not cheap, but equip them with the best gears, arm them with the best weapons, and they could do things more versatile than any robots you could produce.
It seems to me that the answers here miss the point.
If both sides used robots instead of humans, then it would be a game, not a war.
A game based on very specific rules. Robots are purely logical, and follow logical rules.
War is emotional. It is based on hate and vengeance, a desire to actually kill another human. The entire concept of war is about humans against humans. In a bar fight, they use their own fists, they do not sit down at a game console or don virtual reality glasses and fight it out in cyber space.
War is the intent to inflict damage on another human enemy, Life or death. Kill or be killed. Not just virtually, but REAL death. Not a virtual enemy, but a real person.
War is very, very personal.
I am sure this society will have an abundance of robot vs robot games. Even on earth, we have megabot competitions. We have drone competitions. But these are not wars.
'War' is humans killing humans. Visceral. Direct. Bone-crunching. The expiration of a human life. High stakes. No reset. No starting a new game. No 'ooops. I didn't mean to do that' take-back. A finality. The ultimate win-loose. Game over means game over. Fighting for keeps, not for points.
War has as its objective the ending of a human life, not the destruction of a robot.
War is a particularly human thing. It is personal involvement.
Without human loss of life, it is just a game.
There are many mundane reasons you can come up with (e.g. cost) - and there's no harm in including them as additional supportive factors. However, the main reason I personally favour is a rather philosophical one: using humans to fight wars directly is considered to be the best way to discourage war and large scale bloodshed.
This may seem a little counter-intuitive, so I'll give you a real-world analogy.
If you were around (and politically aware) during the Reagan era, you'll have heard of the Strategic Defense Initiative (colloquially derided as "Star Wars"). This was basically a laser-based ballistic missile defence system. The ultimate goal was to have a system so good that it could intercept every inbound nuclear-tipped ICBM (intercontinental ballistic missile) before it could detonate over US soil.
Putting aside logistics and efficacy, mooting such a system seems like a really obvious thing. Given the immense destructive power of strategic nuclear weapons, how can one ever argue against an effective defence against them?
But one of the main arguments against the SDI was that it would destabilise the delicate balance of power between the US and the USSR (the only serious nuclear rival at the time). Because the nuclear assets were considered so evenly matched, the powers were considered to be held in check by the shared concern of Mutually Assured Destruction (or MAD for short, a very appropriate acronym). The main thing holding one country from starting a nuke strike on the other was the certainty of being utterly destroyed in response. But if one country were to develop an effective physical defence against an incoming nuclear strike, then that country would immediately have the advantage and could (perhaps) contemplate a first strike with relative impunity. And (thinking further), even allowing one country to start making serious inroads into such a programme might prompt its chief enemy to jump the gun and try its lucky by launching a pre-emptive strike while things were still relatively "even". That's actually the sort of public political and military debate that was going on at the time.
In the end, the SDI never took off, so all that's academic. But, as mad as it sounds, MAD remains an accepted doctrine.
Now let's see how that can be adapted to your scenario. War should be seen as the last alternative given the lasting damage it does to human societies. Let's say your fictional world is comprised of democratic states (not psychopathic despotic regimes) in a state of delicate military balance with existing military forces more or less serving as an effective enough deterrent against attack by others. However, tensions are always simmering and countries would be quite happy to seize foreign territory and resources if it were feasible to do so.
A country that develops super-efficient and practically invulnerable "killerbots" before any other can immediately start attacking others and annexing territory with relative impunity. In fact, several countries have realised this and started up independent killerbot initiatives, all racing to get the first working prototype robo-army up and running (and killing).
The World Council (like the UN in our world, only more effective) sees this as a very worrying trend. It decrees that the manufacture of killerbots is absolutely outlawed by international convention. Only human soldiers are allowed (and only for defensive purposes). Their reasoning goes along the same lines as the objections to the SDI in our world: human life is considered universally precious (remember - no despotic tyrannies in your world) and the thought of losing their able-bodied young people en masse is repulsive to every nation. That becomes the biggest deterrent to wars of aggression - analogous to MAD. However, allowing killerbot programmes to progress undermines this deterrent. Therefore, the killerbot programmes must be outlawed. And analogous to the second-order thinking related to the SDI that I mentioned, the council also opined that if one country was seen as having leapfrogged over others in coming close to a functioning killerbot, others might immediately attack pre-emptively just to safeguard themselves as they would consider the immediate limited price in blood to be more judicious than a much bigger one down the road (at the white-hot metallic pincers of the killerbots).
So killerbots have been universally outlawed by treaty (you can think of a way of effectively surveilling for treaty-breakers), leaving only human soldiers to fight wars (purely defensive ones, as per the intent of the treaty).
Humans can have a moral/social/honor code. These things are beyond the understanding of a machine into the foreseeable future. It is impossible, once programmed, for a machine to say 'no' to an immoral order. This is for the protection of your own society more than the enemy society.
If you don't want to address the issue yet, you could push off the explanation to some future story. Just talk about how it's already been tried and has failed in the past. Keep the discussion at high level, avoiding the details.
Unless the details actually are the story. If that's the case then all kinds of options exist:
A good reason to not build robots for war could be cost and vulnerability. Not to be crude, but all you need to make more humans is a couple humans and time. Humans can grow/hunt their own food as a community
What you need to make complex war-fighting robots are many industries. You have to get the metal/materials from somewhere, then you have to refine them (if they are metals), then turn the refined metal into robot parts, assemble them, install complex electronics/sensors/radios/processors. That's not even taking into account all the highly trained people to program your warbots or design/iterate them in the first place. Such a supply-chain would be a large vulnerability for an enemy to exploit.
Or to put it another way, building robots costs money/capital and while it may be faster than building human soldiers (once you have the proper industries in place) its much more expensive. With human soldiers it might take longer to "build" them, but it's fairly cheap as the families that raise soldiers generally contribute to your economy as well in the form of labor and tax income and whatnot
From a cynical point of view, the government might not value the life of its citizens (thus the "cost" of a human life is zero) or might even want to reduce the population because of overcrowding (thus the "cost" is negative). Other options would be a right of passage, where each human (or man) is forced to do it or an optional way to increase status (like medieval knights) or gain priviledges such as voting rights or candiature for government offices (I think Starship Troopers had this).
Maybe robots are more intelligent than humans and have higher ethics (like the spaceships from Iain M. Banks) or at least count as citizens and can't be forced.
Maybe humans in that universe have their fear and will to survival removed from their DNA because there are so many of them and don't mind dying and it's a fun adventure for them. They could also be backed up so they can just be revived if they die.
