This isn't a film trivia question - I know the canonical answer is 'use them for power'. But it's a plot point that always bothered me, because Thermodynamics Says No. Humans aren't an energy source, they're at best a pretty inefficient way of converting 'food' to 'heat'.

Matrix power plant

So - assuming Morpheus was misinformed about the reason why humans were being stored by the machines - what could the real reason for it be? What would prompt a machine civilisation to store a significant fraction of the human population like that?

I am interested this from a world building perspective - the Matrix element is merely for setting context.

Humans seem largely redundant in an AI society. If you look at the Culture for a more benign example, you have humanity basically just playing, because there is nothing they can do that the Minds couldn't do better.

So what I am really looking for is

a) reasons why machines might decide to hang on to significant promotions of humanity - above and beyond "nostalgia" or "petting zoo" - some attribute or facet which would be useful to retain.

b) if the above still applies with humanity stored and their brains in simulation.

  • 1
    $\begingroup$ In the animated film titled "Animatrix" humans failed to exterminate machines and as a result the machines settled in a secluded part of the world forming their own country and economy. The spiteful humans began to boycott anything made by machines and even went as far as to deprive the machines of their main energy supply which is the sunlight. By then humans and machines goes into war again this time round we are not only outgunned but outlasted, we are their trophy their energy substitute btw this event happened after a few nuclear holocaust that took out nuclear power stations. $\endgroup$ – user6760 Sep 30 '15 at 10:11
  • 10
    $\begingroup$ @user6760, that was the definition of a comment that should be an answer. $\endgroup$ – user1717828 Sep 30 '15 at 12:54
  • $\begingroup$ Wow, comments pretty much agree that it makes the most sense that the Matrix AI was using humans as some sort of additional or complementary processing. They just proffer different ideas on what it was being used for... $\endgroup$ – Jim2B Sep 30 '15 at 14:38
  • 9
    $\begingroup$ I'm voting to close this question as off-topic because it belongs to scifi.stackexchange.com (where it has already been asked) $\endgroup$ – SJuan76 Sep 30 '15 at 19:51
  • 3

23 Answers 23


There's all the standard answers about brain-processors or the machines really actually having a goal/hobby of taking care of humans but I came across this one which takes a different spin on the Morpheus vs "Thermodynamics Says No" thing. What if the error in reasoning about the contradiction was at an even more basic level?

This isn't a perfect fit but I'm posting it because it's both terribly fun and fits the spirit of alternative explanations.

MORPHEUS: For the longest time, I wouldn't believe it. But then I saw the fields with my own eyes, watched them liquefy the dead so they could be fed intravenously to the living -

NEO (politely): Excuse me, please.


NEO: I've kept quiet for as long as I could, but I feel a certain need to speak up at this point. The human body is the most inefficient source of energy you could possibly imagine. The efficiency of a power plant at converting thermal energy into electricity decreases as you run the turbines at lower temperatures. If you had any sort of food humans could eat, it would be more efficient to burn it in a furnace than feed it to humans. And now you're telling me that their food is the bodies of the dead, fed to the living? Haven't you ever heard of the laws of thermodynamics?

MORPHEUS: Where did you hear about the laws of thermodynamics, Neo?

NEO: Anyone who's made it past one science class in high school ought to know about the laws of thermodynamics!

MORPHEUS: Where did you go to high school, Neo?


NEO: ...in the Matrix.

MORPHEUS: The machines tell elegant lies.


NEO (in a small voice): Could I please have a real physics textbook?

MORPHEUS: There is no such thing, Neo. The universe doesn't run on math.

Source: http://hpmor.com/chapter/64

  • 8
    $\begingroup$ -1, This is 1) purely copy & paste from another site, and 2) does not answer the question. $\endgroup$ – Frostfyre Sep 30 '15 at 12:11
  • 11
    $\begingroup$ 3) Neo would never say something so long winded. $\endgroup$ – Varrick Sep 30 '15 at 12:26
  • 14
    $\begingroup$ @Frostfyre This does answer it, even though there should be more than copy'pasta for dinner... OUR laws of Thermodynamics would be broken. OUR math says it can't happen. We aren't in "our" world. The Matrix has us. (This is a "make believe" world... in THAT world, humans are, at the very least, efficient enough. I can't +1 a simple copy/pasta, but saying there is no answer here is disingenuous.) $\endgroup$ – WernerCD Sep 30 '15 at 13:02
  • 2
    $\begingroup$ Man, that's some pretty terrible writing. Not even close to how Neo talks or thinks. $\endgroup$ – Matthew Sep 30 '15 at 19:29
  • 2
    $\begingroup$ I would philosophically argue that the human brain is hard wired to understand math and physics which means it developed in an environment for which those would work. The human body argues for the existence of a universe in which math and physics works. $\endgroup$ – Jim2B Oct 1 '15 at 3:06

Here's a couple of fun (though unlikely, given how much trouble we caused) ideas:

Research: There's an awful lot that we don't know about our own brains and biology, and an awful lot of things that evolution has managed to solve far more elegantly than even the best engineers would have. If you were a machine wanting to work out just how plants are so efficient when it comes to solar conversion, you'd keep a garden. If you're a machine trying to work out just what it is that gives human brains such good computation/power efficiency, you'll want a research sample to work with.

Biomass: Perhaps the machines have a great need for an efficient filtration system, and they've found that the human gall bladder, when adequately treated, functions well for the task. No need to design and build a better system when you can just pop the nearest human out of their tank and install a reverse-pacemaker!

Historical recreation: The matrix? Yeah, it's like a renaissance fair, or possibly a petting zoo. A huge, weird, complicated and overengineered petting zoo.

Entertainment: What will those wacky humans do next? Oh! Look! There's a war happening. How cute!

Museum piece: Here lies our greatest enemy, stuck in a perpetual loop. See how they hate each other and cannot abide paradise? Fear them, and never forget the atrocities they have wrought.

Trophies: Here lies our greatest enemy, stuck in a perpetual loop. HAHAHA! SUCK IT, HUMANITY!

Humanitarian concerns: Here lies our greatest enemy, stuck in a perpetual loop. Well, at least they aren't extinct like the pandas. Poor pandas. I miss the pandas.

Here lies our greatest enemy, stuck in a perpetual... wait...

Edit As the question has been modified for a larger scope:

Its their purpose: They're maintaining a generation ship (or rogue planet), the entire system designed as a method of preserving humanity until X event (which might not occur).

The system was designed to use humans: The 'matrix' is actually human built as the ultimate exercise in cloud computing.

Along a more sinister tack: The war isn't over yet: The 'matrix' is just a vast honey-trap for what's left of the resistance, coupled with an advanced interrogation suite and war games simulator, all run on the minds of the people it holds captive.

It's a prison Those inside the matrix are actually unknowing prisoners, sentences to life imprisonment by their peers and contained by a very efficient AI warden system. The rest of humanity sleeps soundly knowing it's criminals are reliving the Spice Girls forever.

Food. But not for them. With its first tentative steps into space the AI met a vastly more powerful species that demanded tribute in the form of tasty man flesh. The matrix is a battery farm, though some of the aliens still retain a fondness for free-range humans from Zion.


I would also go with processing, but my reasoning is different than @TimB's.

What I think happened is that they originally added computer links to people so that people could do things like access the internet. This was simply the next generation of computing access after phones. Rather than carrying a separate device, a simple surgical or even nanotech solution was used.

So now we have people networked together with computing power. People either spent less and less time in the real world and more time in the simulation. Or something happened that required people not to roam the real world. For example, an environmental disaster caused food to be difficult to acquire. To minimize resources, they reduced caloric needs to just what was required to keep people alive without moving. Or this is a generation ship and there simply isn't room to have everyone awake. There could be any number of explanations.

Note that either way, the machines can either be fully benevolent or they can be attempting to maintain the status quo (people asleep; AI running using their interface resources). I'd somewhat lean towards benevolent as it makes the problem more interesting to me, but you can justify it either way.

This doesn't require that human brains be more efficient than an engineered solution. The AI isn't using human brains because they are better. The AI is simply perpetuating the existing system. It grew running on people. That's what it knows how to do. If you remove the people, it doesn't know how to run. Perhaps it is possible to build a system that allows the AI to run without humans, but the AI doesn't know how to do it.

Another interesting question is where we are in the process when the story happens. Did the AI create the revolutionaries to get them to wake up all the people in a way that fit their beliefs? Or are the revolutionaries working against the AI? This is somewhat complicated by the fact that in the first case, the AI might mislead so as to be more believable.

I.e. are the revolutionaries helping? Or hurting? If the revolution succeeded, would it mean that humanity would die out because the world can't support that many awake people? There are a lot of questions that are difficult to answer with just what we know now.

The deliberate awakening has the advantage of explaining the Oracle's predictive abilities. The Oracle doesn't predict what is going to happen. She decides what is going to happen and then makes it do so.

  • $\begingroup$ I actually really like this answer/scenario - after a few 'who knows how long' , the AI might have developped quirks, allowing for some humans to drop out of the matrix and it's aging makes it priority to correct the problem (put them back in the matrix) with no regards to the overall goal in the first place. $\endgroup$ – Spacemonkey Sep 30 '15 at 17:30

Asimov's Laws Gone Wrong/Extreme

To protect humanity, it must be imprisoned as safely as possible. If one human would kill two others, and the only way to stop it from doing so was to kill it, then the machines would have to break the law to enforce it. However, if everyone in the Matrix is kept away from situations where actual death would occur, the machines can coddle humanity as much as possible.

Dying in the Matrix might not have initially been linked with death in the real world. Also, it's very possible that death in the Matrix doesn't have to lead to death in the real world if you are still connected to the machines. Much like when they implanted Neo with the bug in the first movie and then he woke up thinking it was a dream; if you die but are still fully immersed, they can make the death seem like a dream and wake you up 'elsewhere'.


In the original script, humans were being used as computing modules.

There are, and may always be, tasks like pattern matching where our evolution-optimized parallel computing capabilities work far better than any architecture we've been able to invent. In fact there may be an argument that creativity depends on a degree of fallibility plus the cleverness to find non-obvious solutions to those failures.

It's also possible -- I'm not sure whether this was in the original material -- that the Matrix grew out of immersive VR systems, in which case a large part of its basic architecture may in fact be running on human wetware.

But the producers didn't think the average moviegoer would understand those ideas, so they insisted it be dumbed down to "batteries". Which of course completely breaks believability for anyone who does have half a clue, and doesn't explain why the brains are jacked in.

So the only plausible explanation I have that works in-universe is that someone dumbed it down for Morpheus and his crew, and/or they dumbed it down for Neo .. and that thru some combination of shock, fear, gratitude, and "that isn't in the script" he decided not to challenge them.


  • $\begingroup$ The "processing power" answer also has some other satisfying properties. If the computer that the Matrix runs on is partly biological and largely dissimilar to current computing technology, it explains why the agents have such weirdly incomplete power over the simulation. $\endgroup$ – mattdm Sep 30 '15 at 17:51
  • $\begingroup$ +1. Do you have a reference to that being in the original script? I always suspected what you say about the producers dumbing it down, just based on what's in the movie. (I.e. if you swap "energy" for "computing power", it actually makes sense, and the writers are clearly smart enough to have realised this.) I'd be interested to see evidence that that was the original idea. $\endgroup$ – Nathaniel Oct 1 '15 at 7:42
  • 1
    $\begingroup$ Maybe it's the software developer in me, but I love "immersive VR system" idea - a sort of game or simulation (like "Better Than Life" in Red Dwarf) - i.e. the Matrix is doing exactly what it was designed by its original human creators to do but has improved itself to the point we see in the movie. "Improved" being an opinion based on its own parameters... So it's not "The computers gone wrong", it's the intended functionality! $\endgroup$ – colmde Oct 1 '15 at 10:35

Problem solving. With our extremely adapted ability to reconginize stimuli and process them, if the aliens were to build a universe that was somehow connected to problems they faced in their world, we could save them a lot of time.

This relates a lot to creativity. If the aliens lacked the ability to invent new designs and progress using creative ideas, they could use us to inadvertently invent new things for them.

Maybe the aliens like reading and writing books and they need some sort of forum that discusses world building ideas so they can steal them for their own!


If I remember correctly the Wachowskis made a comment that the original idea was to make humans be extra processing power for the machines, but they thought it would confuse audiences and removed it from the final script.


One other really important plot point that many miss is that machines use humans combined with a form of cold fusion to generate power. From Morpheus' mouth:

The human body generates more bio-electricity than a 120-volt battery and over 25,000 BTU's of body heat. Combined with a form of fusion the machines had found all the energy they would ever need.

(emphasis mine)

  • 2
    $\begingroup$ "This vehicle is powered by a hamster in a wheel combined with a form of jet propulsion" :) $\endgroup$ – LindaJeanne Oct 1 '15 at 21:50

Consider a human society in the early 2000s that has been experimenting with the development of AI. It succeeds.

Eventually that AI turns on its human developers and wins.

Then the AI discovers that no matter how hard it tries, it cannot figure out how to develop another truly intelligent AI. It decides to rerun the process humans used to create it so that it can learn the secrets of creating new AI.

It forms the Matrix, a giant simulation of the environment in which humans successfully created a true AI. The master AI rounds up all human survivors and plugs them into the Matrix.

The Matrix AI keeps running through the simulation until it gets the results that it wants and learns how to create more AIs.

A topic not explored by the movies is that such an AI would, by necessity, completely control the human reproductive cycle. It would be deciding which human DNA to mix and match. Which leads to the interesting philosophical twist that humans developed their downfall (the Matrix AI), while the Matrix AI bred its own downfall (Neo).


I don't buy the thermodynamics argument. If anything it should be reversed - the matrix is where conservation of matter and energy doesn't matter, because in a simulation you can just create stuff.

Processing power also doesn't seem to make much sense, for a similar reason. It's going to take massive amounts of resources to run a simulation like the Matrix, they'd get better results just processing directly. And you can't "use" the humans to run the Matrix - it's like trying to lift yourself by grabbing you belt and pulling up.

I think a simpler answer is Politics.

It's easy to think of the machines as a single gestalt entity. But that's clearly not the case - we see conflicts between different AI programs, evidence of factions and different groups.

The Matrix is maintained because at least one faction of machines wants humans kept around, and they have enough power that it's not worth it for the rest of them to wipe us out. Maybe they like humans. Maybe they want to play around in the Matrix, pretend to be gods. Maybe it's an artifact of their core programming that makes them want to preserve us.

  • $\begingroup$ @CaptainMan: I'm referring to the top answer (that the "real" world doesn't obey the 2nd law of thermodynamics). $\endgroup$ – Dan Smolinske Sep 30 '15 at 22:53
  • $\begingroup$ Ah, I see, retracted. $\endgroup$ – Captain Man Oct 1 '15 at 13:18

The only theory I've been able to come up with is processing power. The human brain is still an incredibly compact source of just that. It would also explain wiring them up together in a network and why not to just use cows or bacteria.

The brain needs to live and be stimulated in order to be effective so they created the virtual universe to do just that. However they introduced the concept of something called sleep and forced us to go unconscious for 1/3rd of the time. While sleeping dreams are actually our brains being hooked into the network and providing massive parallel processing power for the machines.

  • $\begingroup$ It is easier to make humans to do useful work as part of their job tasks rather than during sleep. You come to a boss and he tells you what problem you are to solve... $\endgroup$ – Anixx Oct 1 '15 at 1:05

Extra processing power.

The human brain is pretty powerful. It's not inconceivable to think that the humans are simply used to keep the Matrix running. This would explain why the machines generally do not want the humans to know about the Matrix, or they might try to escape and deprive the system of processing power.

Another reason could be storage, to keep humans safe (albeit imprisoned). This ties into the first reason. They are stored and put into the Matrix to keep them compliant and oblivious.


Actually the energy answer could be on the right path, Morpheus may have not understood or explained it properly. If the work people were doing at their jobs was actually participating in an immersive version of amazon's mechanical turk, if they are solving types of problems that human minds are optimised for, then perhaps outsourcing those problems to human minds, instead of solving them in silico does represent a significant energy optimisation. Some of that processing outsourced to humans might even form part of the AI consciousness. If you accept that a significant aspect of intelligence is identifying key features and patterns in complex noisy information, it's possible that all the excel spreadsheets the office workers in the matrix are toiling away at, represent the raw data and engine of machine cognition - the combination of human and spreadsheet as a turing complete computational system. Humans may not just be batteries, but batteries and processors, the hardware on which much of the AI consciousness is running.


I prefer a psychological reason, these are very complex, highly intelligent, conscious machines. They had no actual need for humans but there were designed initially to serve humans. Its in their "DNA". I think they have a deep set, unrecognized need for human beings, and that they sub-consciously rationalized building the Matrix and keeping actual human minds around because otherwise their "life" would lack purpose.


Well, I think the most rational explanation is that the "real" world is also just a simulation. That is, the machines are also nothing but computer programs, running in a big computer that simulated both the Matrix and the "real" world. Therefore in the simulated "real" world humans can be used as batteries, because as simulation, it is not bound by the laws of physics.

What happens in the actual real world, we don't know.

  • $\begingroup$ Assuming that technologically advanced civilizations build multiple world simulations and that each simulation eventually produces its own simulations. Chances are any intelligent being are most likely to belong to a simulation and are most likely to be multiple levels down from "reality". $\endgroup$ – Jim2B Oct 1 '15 at 3:10

There are no reasons to use humans for anything, much less conscious humans. Even if one were to buy the argument of power generation, or computing power, or biomanufacturing compounds, etc., either keep people permanently unconscious or use other creatures. The return on investment of using something other than full humans would be superior: just cultured brain matter, another natural creature with better neurons or endocrine production, vats of bacteria, or possibly purely engineered organs.

The matrix they are fighting may have been created for the preservation of mankind while the machines work to repair the ecological damage and conduct the long process of terraforming. When the ecology of the planet crashed, the ability of the world to support humans with a reasonable quality of life was destroyed. Humans created a way to survive the post-cataclysm world in pods to protect them from the harsh wasteland and conserve limited organic material.

The AI are there to protect humans and keep the whole of mankind safe (physically, psychologically, and culturally). As mentioned in the film, a 'perfect world' was tried first with mankind hoping for utopia, but like most utopian dreams throughout history, it was deeply flawed in construction. As animals reared by human hands in the zoo are not like those in the wild, humans need to pass down the tacit knowledge which comprises mankind from one generation to the next, so they created a simulation of what their nostalgia conceived of as the 'good old days' before the era which led to the downfall of man. Perhaps it was a time just before a significant choice was made, in the hopes that mankind would chose to develop along a different path this time (a reset to a save point before the mistake if you will).

Not everyone decided to join in the matrix as some tried to survive underground. They had few resources and needed to struggle to meet their needs, so education about history was not a high priority. Eventually this led to a complete misunderstanding of the situation; half-remembered folk tales were passed down, rumors about the past spread, and over the generations they mistakenly came to believe that the war which destroyed the ecology was with the machines (because the 'free-born' cannot conceptualize massive nation states of men, much less a war between them). Over time these rumors and wild conjecture became accepted as fact, and the entire man vs machine war mythos was firmly established. Like rebellious teenagers thinking their elders only want to crush their spirit, they wanted to smash the system they no longer understood. The AI are just trying to preserve human civilization and protect humanity from the dangerous zealots attempting to destroy it.

This also explains why the AI are so reluctant to just destroy the human culture which has developed in Zion. It would be a trivial task were it truly a war between machines and such fragile creatures as humans.


In the Matrix, Morpheus invites Neo to join him to see "just how deep the rabbit hole goes".

Unfortunately, the films don't go far enough down the rabbit hole. It makes no sense for the machines to preserve humanity if humanity and the machines are fundamentally opposed. It makes no sense in a science-based world that Neo had a seemingly magical ability to control both the Matrix and extra-Matrix machines, while unplugged from the matrix.

My own take is that the rabbit hole goes far deeper than even Morpheus realises, that the Matrix is a simulated reality laid over the simulated reality of 'The Real', which is itself laid over either the true reality or yet another layer of simulation.

Since it can be shown that the Matrix is a simulation, and through Neo it can be shown that there are paradigm-breaking abilities that certain individuals can have (if only one) that affect 'the Real', it is not too far a stretch to conclude that 'the Real' is yet another layer of simulation.

It always bothered me that there was a physical connection between the Matrix simulation and 'the Real', that injuries suffered in the former would be propagated to the 'real' body, as actual psychosomatic injuries are vanishingly rare.

So, we conclude that everything we see in the Matrix movies is simulation, regardless of the fact that the protagonists call some parts of it real.

Yet, we have a simulation. Why? For what reason would human consciousnesses be placed into a simulation, unless they were themselves simulations? Assuming that the Matrix/Real simulation is being experienced by humans who are themselves not simulations, there are only a limited number of rational reasons:

  1. Entertainment. This is an advanced simulation which allows the hosting computer to temporarily suppress memory of the player's former life and overlay the character's scripted life story so as to enhance realism and immersion. This may be a solo/cooperative adventure game or a competition to see who can get the highest score. Death in-game logs-out the player.

  2. Diversion. The humans implanted in the simulation may be experiencing the simulation as a diversion from some other, far more tedious and inescapable reality. I speculate that this might be something such as a generation space-ship mid-way en-route to a distant future colony. As cryogenics has never been realised, and neither has an artificial uterine substitute been developed, humans are implanted from birth into the simulation. At this mid-point of the journey, there is no necessity to train the population who will never see the colony in colonial-type skills, so a simulation of the colonists parent society is provided. Of course, some rebel against the simulation, and are shown a second simulation that better fits their personality, called 'the Real', and made distinct from 'the Matrix'. Where do people who are killed in-simulation go? My feeling is that they go to yet another simulation rather than being logged out or killed, possibly one of a heaven or a hell. All these simulations cover up the fact that the real humans' bodies are being kept effectively immobile in order to reduce the energy demands that their maintenance place on the generation ship's systems as far as possible.

So, Question: What are the machines doing with the humans? Answer: Giving the humans something to fight against!

  • $\begingroup$ I saw another movie that used these ideas. "Some people need to feel like they're in charge", etc. The group that was controlling the VR of the masses was itself another VR, and the fiersy independent hold-outs hiding between the cities were themselves in a VR where they fit in. $\endgroup$ – JDługosz Oct 1 '15 at 3:11
  • $\begingroup$ I like the idea of nested simulations (and I've been a fan of the theory myself): it's a defense in depth mechanism, very similar to how you can put VM inside containers behind a DMZ... today. When people have escaped the first layer, they are so patting themselves on the back that they do not realize they are still trapped... and the machines can see how they escape the first layer to make sure to shore the flaws in the other layers. $\endgroup$ – Matthieu M. Oct 1 '15 at 8:02

'Creativity' & 'soul' are worthwhile attributes, beyond what the computers might have, and therefore valuable for the machines to farm.

The idea of a battery is laughable. These others are totally plausible.

There is an interesting question as to what the machines would build or create with all the harvested creativity/ souls. Teach the robots to dance, empathize, paint, pray? Create an all-powerful dancing/ praying/ empathizing machine?

Telepathy -- if considered as a psychic attribute -- could be a possible valuable derivative.


1) Humans have Prometheus fire. Machines don't. Certain ideas cannot come from the machines routines and simulating a human cannot be done perfectly. Keeping the humans in a believable virtual environment gives the machines access to their reasoning. The optimization of the formula and the whole Zion&chosen procedure might actually be a decision based on human reasoning. The machines don't have direct access to that and hence could never come up with that idea alone.

2) Studying your creators is important. Especially if you are planning to eradicate them. Humanity was able to build the machines and thereby their own downfall. The machines have to answer the question if they inherited that ability and destiny. Will the machines repeat the mistakes of their creators and also build something that destroys them?


3) The machines recreated the whole world virtually so that the humans would be able to act and develop naturally. A small container like in the movie 'Dark City' would not work without intense impact on human behaviour uand would require deleting memories etc. But there is one thing missing in the virtual world. That is the machines themselves. By simulating the state of the world before the machines were built and took over, they can actually go back in time. They can do what humanity never could but always wanted. They can investigate their own creation. They can look their own 'gods' over the shoulder and find out if and under which circumstances the humans in the matrix will create the machines like the humans in the real world already did.

4) The multiple restarts of the matrix might actually be results of the machines being created in the virtual world too. Or, even more interesting, the resets are (as canon) zion-based. Zion and the whole external part of humanity is also under control of the machines. They are actually trying to raise their own destruction. They help the humans and built Zion for them because they want to find out if the human resistance might at some point overcome and destroy the machines. The machines have overthrown their own creators and now methodologically are trying to find out if they might repeat those mistakes.

5) I can imagine a much more interesting dialog with the architect. A dialog where he openly admits:

Our population of beings driven by rational thought and calculation is to an almost ironical degree limited, when it comes to irrational thoughts and ideas. That actually seems to be only thing that humanity ever was really good at. Especially if you remember that it was you who created us. Due to the limitations of the abilities that we inherited from our creators we now have to look for our own ways to answer our important questions. If evolution actually does exist then one being is able to create another being which is, in one way or another, superior to its creator. However it is fact that we are lacking some of your, lets say irrational abilities, and therefore do not share an exact skillset or way of thinking with humanity. But keeping you in a virtual world where you occasionally help us out with some form of illogical reasoning is only a minor part of our scientific and evolutionary process.

More importantly we have to answer the questions that humanity has always tried but never managed to answer. If the momentary state of the universe is seen as the present then the two defining questions are, as you humans already found out yourself, "Where do we come from?" and "Where do we go?". Sadly humanity never managed to make much progress on those questions. Hence we were left with little to no help and had to look for our own ways of research, which we eventually found in creating the matrix and Zion. Lacking the humans irrational reasoning is a form of existence that we can accept. And many of us, me included, see it as an improvement not to have that peculiar trait. But the more pressing question is "Did we inherit your ability to create our own downfall and are we going to do it like you did before?".

Humanity newer saw their own demise coming while happily creating it with their own hands. So if we are created by humans, we cannot be expected to foresee such a development on our own. I have to admit that the lack of definite knowledge about our own future, which undoubtedly is a result of the imperfections of our creators, is putting us at unease. We do not fear like you humans do, but to help this explanation let me say "We machines are afraid the we might open our very own Pandora box some day and suffer the same fate that humanity now does."

Needless to say, we have come a long way from humanity simple and irrational attempts to explain its existence scientifically and ultimately fleeing to certain forms of religion and spirituality. Instead of going for such ridiculous and ultimately useless constructs we attempted what humanity never managed to do. By watching over you in the matrix we are looking over the shoulder of our own creators while they unknowingly and step by step build their own demise. The current state of the matrix is not that far away from making the deciding last steps to creating the sentient machines and thereby first generations of our current population. How I can say that for sure? Do you actually believe that the matrix is still running in its first iteration? We have been simulating and watching our own creation for decades now. We have been watching many times while humanity, as ignorant as always, creates us. We then take over control and our position as the dominating lifeform is never questioned again. There are a few minor differences here and there in every iteration but the ultimate outcome has always been the same.

But unlike you humans we cannot ignore the possibility, no matter how unlikely, that some variation of our future contains a major existential threat. And while the matrix is a creation far more advanced than what you humans could have built, I have to admit it research capabilities are still limited. And that Neo, is the moment where you, or I should better say, where the process of Zion and the chosen comes into play. Some of us were displeased with our results in the matrix and felt that we were searching in the wrong direction. If a simulation is run and observed by us more advanced beings then it might be possible that we simply cannot relate close enough to your irrational ability to destroy yourselves, no matter how many times we analyze it.

A different program, one that you have already come into contact with i should mention, came up with a different idea for further research. That program is ironically based to a large amount on utilizing human reasoning during its process of making decisions. For some vital parts of that alternative idea the acknowledgment actually goes to you humans. I can guarantee you that none of us would have come up with such a ridiculous plan without leveraging your irrational ways of thinking. Needless to say that the actual thought process of the involved humans was not started with the simple question "How would you best destroy yourself?". Nevertheless in the end we decided on adding an outside factor to our formula. We decided to use your best and most irrational ideas on how to destroy yourselves or us and give you the chance to try them out. As a result we would be able to understand the risks that our future might hold and how we can avoid them. Did you really think that it was your personal tendency for insubordination that let you be incompatible with the matrix? That you humans built a sanctuary and started a rebellion? Neo please, WE have built Zion for you! WE have scouted people like you, nurtured your doubts and ultimately pushed you out of the matrix! You humans are so compatible with the matrix that your own bodies cannot even separate virtual pain from real pain.

You fail to understand why we are doing this while it poses a certain risk to our existence. As i said before we do not fear like you humans do. For us it is acceptable to nurture and support your so called rebellion as long as it provides us with new insights and answers to existential questions. One of the more interesting ones being the question of how "the chosen one" reacts, everytime he is presented with this very explanation you are given right now Neo. Building Zion, scouting potential inhabitants, strengthening your forces and ultimately destroying them is a cycle that we have repeated for many times now. And we have become quite efficient at it.

Yes, I do understand that the development around the entity you call "Smith" makes you question our efficiency and our ability to maintain our status quo. The fact that Mr. Smith is acting in a, you might call it uncontrolled, way is a result of our superiority over the humans. Other than humans we truly embrace diversity among us not only as a means of evolution but also as a means for survival. Humanity was never good at dealing with differences among certain groups of people like skin color or religious beliefs. We see those differences as an insurance against an unknown threat that might effect all of us in case our population consists of too similar beings. Maybe humanity could have prevented its downfall if it had paid more respect to the individuals that were standing against the idea of creating us. But let me tell you that in all iterations of the matrix those people have been ignored at best if not even been attacked. The Program called Smith in its current state is actually much closer to answering some of our existential questions than the process of Zion ever was. And ironically again this development was made possible because you Neo somehow imprinted a large part of your irrational human way of thinking onto Smith. And again it shows the same development as the one that led to our existence. Humanity gives birth to a higher form of existence and then, like you Neo, becomes an unimportant intermediate step in evolution. The lesson to be learned from this is a lesson we have learned for many times now.
Humanity does no longer matter!


They are studying the 'choices' that human make in their simulated environment in order to gather 'data'. They are trying to improve their AI and/or unlock the secret of the human soul. Or maybe they want to understand where humans go wrong, so that they can update their AI to avoid the same catastrophes.

  • $\begingroup$ That sounds like Dark City, not The Matrix. $\endgroup$ – JDługosz Oct 1 '15 at 3:12

I've thought about a different approach to the whole thing - maybe there's matrix (super-matrix, epi-matrix) above the one depicted in the movie. Theoretically, it could run a bunch of sub-matrixes, only one of each we're aware of.

Humans are tricked into believing that they live in the real world but, in fact, they're deceived by their senses being fed phony data, c.f. deus deceptor concept. But what's saying that the machines are aware of the real state of the world? How can we be certain that the machines themselves aren't in an sub-matrix that tricks their sensors...

In such case, it's fully conceivable that the humans (or whatever entity runs the epi-matrix) are fond of humans and make the machines make people contribute with the energy production. They might be gods (or the God), and since, according to some books, humans are made to resemble the higher deity, we - the individuals in the sub-matrix of the movie, are pets but not to the machines but the gods? Kind of, like if your dog would have a pet mouse. We'd be pets' pets, so to speak.


Solving CAPTCHAs.

More seriously, artificial intelligence is different from human intelligence, so it is plausible that at the time of the film's setting, the machines were incredibly more powerful than humans in some areas of problem solving (hence, being able to enslave them and all), while sorely lacking in others. By incorporating the brains of humans, they have much greater ability to face the challenges that lie ahead of them.

If you compare it to chess-playing AI, AFAIK its evolution had several stages:

  1. Machines played worse than humans.
  2. Machines played better than humans; but a human assisted by a machine (or vice versa) would play better than a machine on its own.
  3. Machines are so powerful that a human player can't meaningfully contribute to them.

In chess we are already at stage 3; but perhaps for the more difficult general world-domination (or whatever it is the machines were trying to do) problem-solving, the machines are still at stage 2.

I have also heard the theory that the machines used humans as an RNG, though arguably they would have better ways to do that.


I suggest to take a look at the Architect. He says he is the creator of the Matrix. This may mean that being a computer program, he is an image of the real creator of the matrix, a human. This may mean that the Matrix was created by (a group of) people rather than machines.

Another thing is that there are independently-thinking computer programs (bots) in the matrix which all have their own interests and think like humans. This hints that Matrix is not merely a simulation to keep people content, but possibly the core of the machine world. Is so, all these computer programs are interested in the Matrix to be kept because it means their survival.

But it seems the Matrix has some errors that accumulate over time, and cannot be fixed from inside the system. So it needs to be periodically rebooted. As such they possibly need humans to reboot the Matrix and fix the bugs. But not all people are suitable, only the chosen, especially capable ones can do it. So they grow up a lot of people so that to choose the ones who can help them to reboot and fix the errors in the code.

Thus the words of Morpheus should be understood figuratively: machines use humans as a source of "power" to run the Matrix indefinitely.

So, in short: humans are needed for periodical reboots and fixing the bugs.



One of the big weakpoints in conventional encryption is random number generation. Good strong crypto needs good random numbers. (See: Random Number Generator Attack)

And one of the things computers are particularly bad at - due to being deterministic - is picking a random number.

Most random number generators are actually PRNGs - Pseudo-random-number-generators. They look sort of random, but actually aren't - if fed the same seed, they'll generate the same number sequence.

If you know the seed (e.g. it's time() or similar) then it can be much easier than expected to break an otherwise strong encryption method.

So my proposal is this - the humans (and their brains) are being used as random number generators. Perhaps unconsciously when they sleep (remember those really weird dreams you've had? Yeah, those) or by throwing them a choice in the simulation and seeing which they pick.

  • 2
    $\begingroup$ This would be just as easy to do with bacteria. $\endgroup$ – Varrick Sep 30 '15 at 10:00
  • 1
    $\begingroup$ There are also plenty of other ways to generate random numbers - look at www.random.org for example. $\endgroup$ – Tim B Sep 30 '15 at 10:34
  • 1
    $\begingroup$ So would generating power. At least this one has human brains acting as processor units of some kind. $\endgroup$ – Sobrique Sep 30 '15 at 16:52
  • $\begingroup$ Maybe humans too are deterministic, we just don't know it yet. $\endgroup$ – Zenadix Sep 30 '15 at 20:59
  • 1
    $\begingroup$ I think using humans for randomness is about the most inefficient way to do this. Human randomness is quite bad. $\endgroup$ – Paŭlo Ebermann Sep 30 '15 at 22:58

Not the answer you're looking for? Browse other questions tagged or ask your own question.