Why is the Matrix a collective? That is, why are humans in the bioelectric harvesting fields part of a shared simulation?

Why is this better than having each subject jacked into his or her own simulation? It is clear from one of the training constructs that the Nebuchadnezzar crew uses ("The Woman in Red") that people can be simulated, and so it seems that a simulation designed for an individual could be populated by virtual characters as the machines see fit.

Also, discussions in other questions on this site have made it clear that

"redpills" are allowed to escape from the Matrix, so that their threat to the Matrix can be contained.

It seems that if subjects had their own individual Matrices, then this threat would be rendered moot and the machines would be able to mothball the One altogether.

8 Answers 8


Take a few assumptions for granted, a matrix meant for a single human would require a ton of programs willing to be NPCs, with both reasonable and unreasonable responses scripted, and hope the PC does not reject it. For any PC that is in a city scenario, that means a bunch of unique looking background NPCs, and a significant number of interact able ones. At that point, if they are not AI, the PC could grow suspicious. If they are AI, the program would have to be willing to do it as we know many programs have free will and can even rebel against the machine source.

By networking the humans together, they provide their own interaction, greatly reducing the processing and programming requirements, and minimizing the number of sentient machines or programs needed for the matrix to exist.

Given that, the machines weren't purely evil in their intents. They wanted to placate the humans, as far as giving them a utopia. They understand that we are herd (or pack) animals and need socializing to thrive. Networking the humans was as much a means as an end.

That said I still think there are multiple matrices running concurrently, like a MMORPG does, but there is no evidence for that aside from Smiths "entire crops lost" statement. It implies that groups of humans physically nearby were lost at once, instead of the entire matrix collapsing.

As for the rebels having the training Sims, they were scripted. A bunch of people moving in one direction, no interaction. Nothing close to a realistic human conversation.

Mouse: The woman in the red dress? I designed her. She, uhm, well she doesn't talk very much, but, but if you'd like to meet her, I can arrange a much more personalized meeting.

Switch: Digital pimp, hard at work.

  • 1
    training Sims, they were scripted Yeah... as if the woman in the red dress would ever actually look at you like that.
    – WernerCD
    Commented Jun 23, 2015 at 19:55
  • 4
    She will if her pimp arranges it
    – user16696
    Commented Jun 23, 2015 at 20:25
  • 12
    I took the "entire crops lost" comment to be about proximity in time, not space. "All the humans from that year" were lost, not "all the humans from that field."
    – Nerrolken
    Commented Jun 23, 2015 at 22:15
  • This is nice work, @cde. I'm convinced. :-)
    – Praxis
    Commented Jun 24, 2015 at 17:52

Doing so would defeat the entire purpose of the simulation. As stated by the Architect, they had tried making the simulation a Utopia, making the humans happy. However, this failed. The result being a new Matrix, designed to be life-like, and ultimately designed to produce The One.

From IMDB, the Architect has this to say on the subject. Basically, the original Matrix failed because humans were not given a choice.

As I was saying, she stumbled upon a solution whereby nearly ninety-nine percent of the test subjects accepted the program provided they were given a choice - even if they were only aware of it at a near-unconscious level. While this solution worked, it was fundamentally flawed, creating the otherwise contradictory systemic anomaly, that, if left unchecked, might threaten the system itself. Ergo, those who refused the program, while a minority, would constitute an escalating probability of disaster.

So, they were given a choice. Then, it was discovered that a small percentage would still reject the Matrix. The solution was to wipe everything out and start over.

The function of the One is now to return to the source, allowing a temporary dissemination of the code you carry, reinserting the prime program. After which you will be required to select from the matrix 23 individuals, 16 female, 7 male, to rebuild Zion. Failure to comply with this process will result in a cataclysmic system crash killing everyone connected to the matrix, which coupled with the extermination of Zion will ultimately result in the extinction of the entire human race.

If they made it so that the humans were not linked together, it would have been difficult for those in Zion to get into the Matrix and recruit The One. Without The One, the machines would lose their power source.

I would argue that "red pills" are allowed to leave to facilitate the creation of The One, not to minimize the risk.

  • 1
    The same choice could be allowed in a single person matrix too...
    – user16696
    Commented Jun 23, 2015 at 17:09
  • 1
    @cde sure, but the Zionites getting into the Matrix to extract people was crucial. It would be impossible if only one human was allowed in at a time. Commented Jun 23, 2015 at 17:16

Same reason websites can be on shared servers: save resources.


cde's answer is much better than mine, but to clarify what I mean by saving resources I'll use a quote from him:

By networking the humans together, they provide their own interaction, greatly reducing the processing and programming requirements, and minimizing the number of sentient machines or programs needed for the matrix to exist.

If each person was in their own matrix, each matrix would require the same programs across them all. If one program needed to be updated, deleted, etc that is each individual matrix that would need this work done. Even if this didn't necessarily take up more resources to run, the amount of resources needed to maintain this kind of system would be greater.

  • The analogy might be better as user accounts instead of servers. In other words, one giant server with many different, ring-fenced accounts, for security purposes.
    – Praxis
    Commented Jun 23, 2015 at 16:22
  • 2
    If each person had their own personal Matrix, how would socialisation happen? Without socialisation I'd guess they'd reject it (too much like the Hell Matrix) and it'd crash. So to control this you'd have to have a vast amount of totally realistic AI characters in each persons Matrix. Big resource drain. On top of that, for each person you'd have to simulate an entire city (minimum). So for each car in the normal Matrix, you have to store 7 billion cars in the individual simulations. And that multiplication happens for EVERY item in the simulation!
    – DavidS
    Commented Jun 23, 2015 at 16:32
  • 1
    @DavidS that seems a bit excessive. Any given individual wouldn't need to know where most cars or objects would be. Ever play GTA, random on demand car generation would be good enough for x miles around the person.
    – user16696
    Commented Jun 23, 2015 at 18:38
  • 1
    @cde But you're well aware that GTA is cheating - in fact, it doesn't really even try too hard to conceal the fact. You don't expect that in the real world - when you run after that guy on a bike, you expect to still see him on the next intersection. In GTA, he just disappears - not good for immersion in something that's supposed to be the real world :D
    – Luaan
    Commented Jun 24, 2015 at 8:09
  • 1
    +1 But you could expand on why saving resources is important enough to put up with the complexity of the whole 'One' scheme... As the matrix was their power source they would want it as efficient as possible. Every bit stored, every calculation made has a real energy cost. Without resource sharing, more of the energy produced by each individual would have to be consumed by their own simulation decreasing yield. At some level of detail, the energy cost of running a single simulation will exceed the production of the occupant making the whole thing pointless.
    – Mr.Mindor
    Commented Jun 24, 2015 at 19:51

The only answers are:

  1. They tried it and it failed or was somehow less desirable than what we are presented with.
  2. They have not tried it yet because the current is working fine as-is or believed to be better.
  3. They have not thought to try it.

The third answer is a weak one and requires no explanation, so I will pass over it. The other two are similar in that the current implementation is at least possibly better.

My answer will expand on Dave Johnson's and cde's answers.

The only other Matrix we specifically know about is from The Architect, here is the quote again, emphasis my own (Edit 1: Everything before the [...] was added in this edit.) :

The first Matrix I designed was quite naturally perfect, it was a work of art - flawless, sublime. A triumph equaled only by its monumental failure. The inevitability of its doom is apparent to me now as a consequence of the imperfection inherent in every human being. Thus, I redesigned it based on your history to more accurately reflect the varying grotesqueries of your nature. However, I was again frustrated by failure. I have since come to understand that the answer eluded me because it required a lesser mind, or perhaps a mind less bound by the parameters of perfection. Thus the answer was stumbled upon by another - an intuitive program, initially created to investigate certain aspects of the human psyche. If I am the father of the matrix, she would undoubtedly be its mother.


As I was saying, she [The Oracle] stumbled upon a solution whereby nearly ninety-nine percent of the test subjects accepted the program provided they were given a choice - even if they were only aware of it at a near-unconscious level. While this solution worked, it was fundamentally flawed, creating the otherwise contradictory systemic anomaly, that, if left unchecked, might threaten the system itself. Ergo, those who refused the program, while a minority, would constitute an escalating probability of disaster.

(Edit 1: The striked through text is incorrect. The text before it and the was added in this edit.) The Architect describes various previous Matrices he created, specifically a "flawless" and "sublime", "naturally perfect" one and one "reflect[ing] the varying grotesqueries of [human] nature". These are both described as failures. This "solution" above is specifically referring to the paradise-like Matrix made by The Oracle that didn't work out. The idea is, as I understand, that something in the human mind understood that perfection (Edit 1: and horrible imperfection) was impossible and something was wrong.

My belief is that the machines/programs while able to closely resemble human behavior are not 100%, and that a Matrix where every other being was a program would trigger a similar response from too many people. (Edit 1: This is supported in part by The Architect's inability to create an adequate Matrix. Of course it begs the question that if The Oracle is able to do it then surely it's at least possible to resemble human behavior (the scene where the program says that love isn't a human emotion, it's a word that implies a connection for example), but as stated only 99% accepted the current program implying it's still not perfect. It's not a big leap to assume that more interaction with the machines and programs would increase this rate and it would become a bigger and bigger issue.)

  • Great answer and welcome to the site. +1.
    – Wad Cheber
    Commented Jun 23, 2015 at 22:31
  • A genuinely thought provoking answer. Commented Jun 23, 2015 at 23:10
  • 4
    No, the Architect's quote is referring to the choice-based Matrix with the path of the One. He had already talked about the Paradise Matrix and Nightmare Matrix earlier in the conversation.
    – Null
    Commented Jun 24, 2015 at 5:03
  • @Null you are right, it's been a while since I saw the film. I found some more of the quote and will edit accordingly. (The quote I found is actually better for my point.) Commented Jun 24, 2015 at 15:00

I'm not terribly familiar with Matrix lore, but I know that human will can modify the Matrix. This leads me to believe that the humans' own brains are being used to at least partially handle the processing workload of simulating the entire world. It stands to reason that the resourceful machines use humans not just as energy sources, but also as parallel processors to create the world they're contained within. If that's true, then of course you want all of the human minds working on the same simulation in parallel. Not only does that reduce the cost of the entire simulation, but it also reduces each individual human's control over the entire simulation. If each human had their own simulation and their brain was used for the processing power behind it (e.g. as is likely the case on the Nebuchadnezzar's private simulations), then each person would have far more control over their simulation and would be far more likely to notice it change on accident.

In other words, all people in one simulation is more efficient and stable than individual simulations.

Check out this Rick and Morty episode to get an idea of what I mean.

  • The training sim contradicts this. They are programmed, and an external operator is needed for changes.
    – user16696
    Commented Jun 23, 2015 at 20:49
  • @cde Each of the training simulations involved the user bending the simulation to their will in real-time. The jump simulation, for example, depends entirely on the jumper's ability to see through the simulation and will their body to make the impossible jump. A programmer is not involved in that process.
    – talrnu
    Commented Jun 24, 2015 at 4:41
  • @talrnu I would say that that is not an example of the participant bending the simulation, rather bending themselves. Commented Jun 24, 2015 at 17:36
  • @CaptainMan That's a good point, but I'm inclined to disagree with your interpretation of it. The very act of describing the user's ability to manipulate the simulation by manipulating oneself (e.g. "It is not the spoon that bends, it is only yourself") only further supports the theory that the user's own mind is involved in producing the simulation.
    – talrnu
    Commented Jun 24, 2015 at 18:15

If AI cannot become truly sentient, like humans, then a human in a simulation will figure out that they are the only true human. Additionally, I've heard that in the original story the computers enslaved humans not for their energy, but for their unique capabilities that the AIs could not replicate. Consequently, there is more benefit for the AIs by combining the humans and having them interact than by separating the humans.

  • But is it not established that AI is indeed truly sentient?
    – Ghanima
    Commented Jun 23, 2015 at 21:01
  • I don't think so, b/c at the end of the second movie, Neo says that what distinguishes him from the Smiths is his free will. If the AI were sentient it would have free will.
    – yters
    Commented Jun 23, 2015 at 21:06
  • I'd like to say "what second movie" but ok ^^. But do the architect, the oracle, the merovingian and so on not have free will (despite the latter believing in causality not choice)?
    – Ghanima
    Commented Jun 23, 2015 at 21:11
  • 1
    This doesn't sound too strong as an argument - your human would simply assume everyone around them is an idiot, and they're a genius. Remember, their whole environment is simulated - from their very childhood. I have many friends who feel that way - is that a proof we're in a simulation? :P
    – Luaan
    Commented Jun 24, 2015 at 8:13
  • 1
    Perhaps they are right. What is the Matrix? The computer simulation in the movie may be a metaphor for something real.
    – yters
    Commented Jun 25, 2015 at 2:05

I don't know of a canon answer to this, but I have this theory...

There is apparently a minimum number of processor cores (read: human minds) that must participate together to produce enough processing power to simulate an entire planet and have it be believable. The simulation system, for example, could only generate a small city while two people were within in it, and not the entire planet. It was much smaller in scale. A sole human mind couldn't run the entire Matrix simulation in their mind and have it remain 100% believable, or it would fry their brain completely from having to work at 100% capacity all the time.

If only one person were needed, the machines could just choose one person at random, jettison the entire human race, and start over each time the anomaly occurred, and yet just over 20 people were required by the system to seed the next generation. This is clearly less than the calculations for what we have predicted would be needed for genetic diversity for perpetuation of the human race, so it's reasonable to surmise that the machines know how to manipulate the gene pool, and therefore only need the power of those chosen minds to run the full simulation temporarily while they replenish their stock.

You also couldn't share the processing power if everyone were in their own personal grid. Also, humans are known to have psychotic breaks when placed in isolation for long periods of time. For most humans, this wouldn't matter, because they'd chatter on happily to that weird person at the bar that never speaks back, but that small percentage that are attuned to the fact that they are, indeed, inside a simulation, this would drive them crazy, possibly infecting other nodes or crashing the system faster. It's much safer to simply pool resources together and not take the chance.

It's not directly stated this way in the movies, but it's plausible. There's a short story about how an asteroid hit the Earth, and destroyed a tower containing 200,000 humans that were simulating London. London literally disappeared while the processing power rerouted data to get to simulating London again. That was mentioned on Why Did the Machines Even Bother With a Matrix?.


We do not know exactly how efficient the Matrix is as a power source, but we do know that the Machines are proud of its efficiency. Near the end of the second film, the Architect makes comments about this at are the closest we ever see any Machine (other than Smith) come to outright bragging.

We do not know exactly how much power the Matrix can get from a human, but we do know two things:

  • It is not infinite. If the Machines could get infinite power from humanity, then they would not bother with the rigamarole of farming millions of humans. They could do it with only a couple thousand, only one of which would be needed to provide power at any given time (the rest would provide redundancy, replacements, and a sustainable gene pool).
  • They can get more out of a human than is required to account for that human's presence in the Matrix. Otherwise, humans wouldn't be a worthwhile power source.

We don't know the specifics of the state of AI technology in the time period in which the Matrix takes place, but it's safe to assume that AI probably doesn't consume computing resources (including power) at a lower rate than our current attempts at it in the real world. And in the real world, our best AIs are quite expensive indeed.

Now, consider that humans, as a rule, crave social contact. They need to live in an environment where other people exist. If they don't, they break down: something which, in the world of the Matrix, might cause them to reject it. Needless to say, this would cause problems for the Machines.

Therefore, the Matrix must provide some kind of social contact for its inhabitants. Theoretically, the Machines might be able to simulate social contact, but that would require one AI Program per simulated person. An average human would need several hundred of these simulated people in order to feel comfortable. But if AI is expensive, then running this many Programs -perhaps 3-400 per human- is almost certainly far more taxing on power than the Machines would want to sustain. It might even require so much power that humans would no longer be viable as a power source

A more efficient approach is to use humans not just for power, but to provide each other's social contacts, at a rate of one simulated person per actual person. Then the Matrix gets the social functionality it needs "for free"; the humans themselves provide it, rather than the Machines. Since the Matrix does not have to run "social AIs", it has that much more power available for the Machines themselves.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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