In Plague-inc the most typical strategy is trying to infect everyone, while staying hidden by not harming infected people (when a virus, bacteria, fungus is found, people start working on the cure). Once everyone in infected the epidemic starts to kill.

However, because it is a game, it has some "advantages":

  1. A human player:
    He knows that he has a strategy to execute. The strategy may be quite strange, like infect everyone without causing them any harm so 'they' cannot find him. Once everyone is infected he evolves lethal symptoms to kill. The effect of this strategy is to maximize his goal: infect, kill. Stay hidden while infecting 'them' would give him a head start, as 'they' are not working on the cure until he gets into the 'kill' phase.

  2. Full environment data is exposed.
    The player knows how many of 'them' are left. He knows if anybody or a region is not infected yet.

  3. A specific goal: to kill all people, not to survive (as a virus). Kill all hosts also means a virus (or anything else) would not survive, because they are parasite.

In reality, those advantage may not be fully available.

Thus, I would like to ask if there are any realistic explanations for the case of a disease (virus, bacteria, fungus) executing the strategy "infect all before you start killing them". In this case there are no human (or alien sentient) monitors directly.

  • 16
    $\begingroup$ How would a virus (or whatever) know that "all" are infected before turning lethal without any intelligence behind it? Would you be ok with an incubation period longer than what it would take to infect everyone? $\endgroup$ – Raditz_35 Jul 3 '17 at 15:10
  • 10
    $\begingroup$ How would virus suddenly modify itself to become lethal, after infecting all people? Local mutation in one patient will merely cause new strain to be created, strain which has literally no advantage in fact that other strain has already infected everyone, in fact, competition might make infecting others harder. $\endgroup$ – M i ech Jul 3 '17 at 16:51
  • 20
    $\begingroup$ Just to state on the record, killing the host is the worst thing any parasite can do $\endgroup$ – Andrey Jul 3 '17 at 16:54
  • 5
    $\begingroup$ Rationally, only an engineered disease should act like this. A natural disease acting likely this is so unlikely and evolutionarily contrary that anyone confronted with it would assume that it was engineered. $\endgroup$ – RBarryYoung Jul 3 '17 at 20:27
  • 2
    $\begingroup$ FYI you're actually better off not staying under the radar for too long if you plan to achieve a quick win at games like Plague Inc. You want symptoms that increase contagiousness and disable your recipients' potential to find a cure. $\endgroup$ – Denis de Bernardy Jul 4 '17 at 7:29

23 Answers 23


An infection suddenly becoming lethal after infecting a majority of the population is pure fiction.

The closest real world analogue are diseases that have an asymptomatic infectious phase before they start harming people. Unlike in Plague inc. where every infected person suddenly gets ill, real diseases don't coordinate between people.

  • 1
    $\begingroup$ it's ok to be fiction. But it should be realistic as possible. $\endgroup$ – Haha TTpro Jul 3 '17 at 15:45
  • 63
    $\begingroup$ @HahaTTpro You asked for a reality-check. Spontaneous coordinated mutation of an entire strain of bacteria doesn't check out. $\endgroup$ – sphennings Jul 3 '17 at 15:46
  • $\begingroup$ @sphennings I know little about diseases, but is it feasibly to develop a disease that start out mostly benign, but mutates really quickly? I'm not saying the developer knows exactly when it will mutate to be lethal, or even necessarily how. But if they have the disease target, say, the lungs (lets say it starts out as just a cough), and they make it mutate fast enough that it can spread to someone who's already had it (like the common cold, but quicker), could they guess that soon enough it'll mutate to be harmful to the lungs? Even lethal? $\endgroup$ – Lord Farquaad Jul 3 '17 at 21:24
  • 24
    $\begingroup$ Diseases mutate all the time. That's not the issue. When a disease mutates it starts with one bacteria in one person. You effectively have a new strain with a new patient zero. One of the mechanics of Plague Inc. is that you can change the characteristics of the ongoing infections simultaneously in everyone who is infected. That is what was asked about and it is impossible. $\endgroup$ – sphennings Jul 3 '17 at 21:30
  • 3
    $\begingroup$ @ceejayoz I've thought about playing with that solution from time to time. However, telomeres effectively count cell reproductions, not time. To measure time this way, you would need to have very regular cellular reproduction, independent of temperature variations. It's not very feasible, though the idea of the pathogenic equivalent of a Swiss Watch does have a certain elegance to it! $\endgroup$ – Cort Ammon Jul 5 '17 at 17:16

The most egregious break from reality in Plague-Inc is that every pathogen around the globe suddenly mutates from harmless to lethal at the same time. This is not how real-world diseases work. When there is a mutation, that's actually a new disease which evolves locally in one place and then spreads from there. Anyone infected with the old strain won't experience any of the symptoms of the new strain until they also get infected with that one.

If you want a humanity-exterminating killer disease, create a disease which is:

  1. highly contagious
  2. has a very long incubation period without any symptoms
  3. is deadly when it actually breaks out

think: airborne HIV.

  • 6
    $\begingroup$ That's both terrifying and disgusting, given that HIV is an STD...(jk) $\endgroup$ – cst1992 Jul 3 '17 at 18:54
  • 23
    $\begingroup$ @cst1992 Sure it's terrifying, but I don't understand what's disgusting about it. A disease where the primary transmission vector is no longer sexually is no longer a sexually transmitted disease. Airborne HIV would be HIV minus everything sexual about it. $\endgroup$ – Philipp Jul 3 '17 at 23:58
  • 7
    $\begingroup$ @Philipp I think "(jk)" usually means "I am joking" on the internet. $\endgroup$ – Tomáš Zato Jul 4 '17 at 8:39
  • 2
    $\begingroup$ @Baldrickk The problem with such symptoms is that they are a warning sign someone might be infected. When the disease finally gets discovered, anyone with just a slight sneeze will be quarantined. When there are no symptoms whatsoever, people might not even suspect that they are spreading the disease even after it was discovered. $\endgroup$ – Philipp Jul 4 '17 at 15:18
  • 2
    $\begingroup$ @Philipp true, but you do need some method of transmission. The cough wouldn't necessarilly need to be a bad one, just the occasional dry throat for example. I guess you could exhibit no symptoms, and wait until another illness is picked up, e.g. the flu, which does cause symptoms and have that spread your illness. Patients will likely be tested for the obvious cause - the flu - and test positive for that, meaning that they will likely not be tested for other illness - at least until it becomes well known and compulsory screening is implemented due to the risk. $\endgroup$ – Baldrickk Jul 4 '17 at 15:27

Quorum sensing.

This concept pretty much is defined by the question. An organism behaves in one way at low population densities, but then switches to an alternative mechanism once they reach higher densities. This is usually done using a chemical trigger upon reaching high population densities.

While this system only works on relatively short ranges in known biological systems, there is no reason that this mechanism cannot be expanded to larger systems.

For example, an infected host can constantly send out large amounts of spores, which do a double job of infecting new hosts and indicating that it is now infected. When infected hosts detect a critical threshold of spores, they start expressing the lethal set of genes, which results in them killing the host.

Reality check

However, it must be noted that this system is extremely unlikely to evolve naturally, and would likely have to be engineered. No monitors are required after the initial engineering phase, however.

This is because killing the host is evolutionarily detrimental to the pathogen's fitness, and there is a huge selection pressure for mutants that violate the quorum sensing and do not kill the host. An engineered pathogen will have to include elements that inhibit the selection pressure.

  • 3
    $\begingroup$ In Halo, a signal system like this could steer the behaviour of the Flood in the more primitive stages. $\endgroup$ – MauganRa Jul 4 '17 at 13:34
  • 1
    $\begingroup$ locust do this to determine when to swarm. $\endgroup$ – user32463 Jul 5 '17 at 22:45

Consider a parasite with two reproductive strategies. Success is fitness: the number of progeny produced. The parasite does not know the big picture worldwide.

A: A parasite which runs its host into the ground is a dead end after that. But in the short term a lot of progeny will be produced by the dying host which equals a lot of genetic fitness.

B: A parasite which limits reproduction and keeps the host healthy will maximize opportunities for transmission. But in the short term it does not have the fitness of one which maximizes reproduction in a single host.

In a circumstance where transmission is easy (for example a crowded refugee camp or a dorm) you could use mode A without much fitness disadvantage - the ease of transmission means the short window before your host is moribund means you can transmit and run your host into the ground both.

In a circumstance where transmission is more difficult (spread out population) you want to maximize the period when your host is mobile and well, to facilitate transmission.

If there is a point at mode B (stealth) where additional transmission is unlikely because all local hosts are infected, then I can imagine an individual parasite which switches modes and runs the host into the ground with rampant replication. When Mode B is unlikely to provide additional benefit so an individual parasite (or a single in-host population) that switches to mode A gets the most fitness. The possibility for making this switch might be built into the genome of the parasite. There would be a balance of selective pressures: for an individual to switch early and give its own progeny the runaway advantage vs staying low level and giving your progeny the contagion advantage.

HIV does not plan ahead but this switch is more or less how it turned out. A large asymptomatic population facilitated transmission (B type) with persons dying (A type) only later in the course of the illness. Many other viruses use this approach - maintaining a B-type stealthy population reservoir in a well population that can then serve as the nidus for an A-type epidemic when circumstances are right. This is how the viral diseases that ravaged New World populations got to the New World. No-one would put a person dying of virus on a ship for a long ocean voyage - these people were virus carriers.

As re every single person in the population suddenly becoming ill (as apparently is the case in Plague Inc) consider again HIV. The number one cause of death in persons with advanced HIV is pneumonia. The death rate for flu among persons with HIV (pre-HAARt) is between 100 and 1000x that of the general population. If you posit two hits - a set and spike type deal - one could have all the infected population fall ill and die because of a second factor which comes through.

  • 2
    $\begingroup$ This may be the most realistic answer to this question but still does not achieve the goal of the questioner. This would produce a behavior change on a local scale, but not a global one. I don't think your description of HIV is a little off. The symptoms of HIV come after time and buildup of virus. There is not a change in the virus's behavior when everyone around the infected individual is also infected. $\endgroup$ – BobTheAverage Jul 3 '17 at 17:19
  • 1
    $\begingroup$ The New World wasn't that the virus changed - just that it encountered a different population; with zero resistance. $\endgroup$ – Martin Bonner Jul 4 '17 at 5:38
  • $\begingroup$ I'm not sure whether the late phase AIDS actually enhances chances of the virus spreading, since it remains chiefly an STD/bloodborne disease. Do you have any evidence that it actually enhances viral spread? $\endgroup$ – March Ho Jul 5 '17 at 9:32
  • 1
    $\begingroup$ @March Ho - AIDS transmission correlates with viral load. Here is one article. nejm.org/doi/full/10.1056/… But HIV does not plan. Endemic spread turns to epidemic spread when individual organisms that have behaviors conducive to epidemic spread produce more offspring than individual organisms that stick to an endemic spread pathway. $\endgroup$ – Willk Jul 5 '17 at 13:45

I'm thinking yeast infection.

What if the disease was a type of fungus. It's perfectly viable in the normal environment, including the air and most surfaces, and generally lives comfortably on the skin or inside an animal host. It's ubiquitous and nobody looks twice at it - except maybe when that sandwich you left in the fridge starts getting fuzzy. If it's part of the natural body symbiont flora in the digestive system, mucus membranes or skin your body won't have a strong immune response. It's impossible to get rid of due to its prevalence in the environment.

We've got ubiquity and stealth, what sets off the mass extinction? A drop in a critical nutrient or energy source makes the humans (and other animals) the best option as a seed bed to keep the fungus alive. It doesn't particularly care about the life of the host, it's just food. Killing the host quickly would eliminate the immune response and make the food even easier to get at.

I'm thinking necrosis of the skin and exponential growth in the mucus membranes are the primary symptoms. Yeast infection and thrush at a grand scale.

There are some fun fringe scenarios too.

  • The fungus likes mucus membranes in the mouth and nose. Rapid growth plugs the airways, suffocating the host. Growth in the lungs would work as well.
  • Gut flora go wild, starving the host as they use all the nutrients.
  • Byproducts are toxic to the host. If it is in the lungs, a fermentation process would release carbon dioxide, suffocating the host. Maybe it produces something like cyanide?

Even if somebody does survive, the whole ecology is going to be devastated. Most animals will be dead from the same infection, and it might cause enough growth and mutation to start going after plants.

I have no idea how realistic this would be, but it fits the model.


An alternative to the resource shortage surge in growth, you could try the strategy of the periodical cicada . They hibernate for a large period, usually a prime number of years. They then burst from the ground en mass and breed like mad. Any predator is rapidly overwhelmed trying to keep up with the sheer volume of prey. Make the fungus experience a rapid growth cycle periodically. If the period is long (say 97 years), people would remember the last time time as a rather annoying bit where food spoiled faster, and a lot of ponds got really scummy. But this time, something in the modern environment provides way more energy than the previous cycle and you get the huge problematic attack on people.


As others have pointed out, it won't work like in Plague Inc: If a relatively harmless virus spreads through the population, and somehow evolves in one place to become lethal, then the lethal version has to spread again, the harmless viruses will not suddenly all change to become lethal.

However, it is possible to have a virus that first infects everyone and then, later, kills many people. It probably won't infect 100% of the population nor kill 100% of those it infects. But it could affect a large enough part of the population to end society.
The evolution of such a virus without a human designer is very unlikely, but not impossible. Here are some options for how it could work:

Long Term Effects

The initial infection with the virus is harmless and only causes whatever symptoms are needed to spread the virus. But in the long term, either the infection damages the body so thoroughly that it becomes difficult to survive, or it causes some secondary effect (through toxins, destroying tissue, overstimulating the immune system...) that kills the host.
It could take many years between the initial infection and the secondary damage, so that people won't discover the connection and won't take drastic measures to avoid the virus's spread before it's too late.

A real-world example is HIV. The reason it didn't infect everyone is because infection can only be spread by intimate contact or medical procedures. So humanity as a whole had time to find medication that can render the infection relatively harmless. If the infection was spread easily by casual contact, things would be quite different.

In this scenario, it would almost have to be a virus, because bacteria or fungi are generally easy to kill and therefore are unlikely to remain active in the body for very long. Bacterial infections generally aren't as contagious as viral infections, either. However, they could in principle kick off some other process in the body, that ends up killing the host when the bacteria are long gone. For example, they could produce harmful prions somewhere in the body, that cause nerve degeneration years later. In this situation, the connection between the infection and the deadly effect would be especially hard to establish.

A Secondary Trigger

Again, the virus infection itself is harmless, but the virus stays active in the host's body for some time (months or years) and ends up killing the host once some secondary stimulus is introduced. This would have to be something that happens to everyone, but not often. It also needs to be something that can't easily be avoided: If people start dropping dead after swimming in a pool or getting anaesthesia, those things just won't be done any more.
Maybe the virus has a deadly interaction with the common cold. Or it makes the body unable to tolerate heat, cold, summer sunlight. Or it causes a deadly allergy to some common pollen: All these things won't show up for a few months if you time it right. Various science-fiction triggers can also fill this role.

For this scenario, your infectious agent could be anything. Virus, bacteria, even some environmental chemical (though you'd have to explain how it spreads to everyone). And since the trigger can be something that everyone is exposed to at roughly the same time, you could have the sudden effect that you want. However, the death rate will probably be lower than in the first scenario, because there will always be groups of people who manage to avoid either the infection or the secondary trigger.

There's one other thing that makes reality different from Plague Inc: Just because people notice that a virus exists and may be connected to some deaths, that doesn't necessarily mean they'll find a cure in any specific timeframe. So your virus doesn't have to be undetectable to have the intended effect, it just needs to be able to spread in a hard-to-avoid manner.


In order to become lethal universally in short order, there needs to be some sort of universal stimulus. The mutation for lethality must also be somehow beneficial to the virus in the short term by allowing it to find new hosts, or overcome a new obstacle. Consider the following scenarios:

  1. A rapidly adapting, highly communicable disease that is naturally parasitic of humans is normally not deadly to them (think common cold). If someone were to develop a medicine that suppresses the virus and distribute it worldwide, the virus will adapt to overcome this. If the simplest mutation to overcome the new, global medicine also happens to be lethal, the virus could essentially spontaneously mutate in every region of the world and kill all of its hosts before it had time to scale back.

  2. A new alien pest (not necessarily other-worldly, but maybe previously isolated) like a rodent rapidly spreads around the world. The virus, which normally infects humans, has a simple mutation that will allow it to infect these new pests, but will start killing humans. The upgraded virus can spread from pest or humans, to pest or humans, so it takes off and replaces the old virus, and probably very similar versions of it emerge in many places because the changes are so advantageous. Essentially the virus changes preferred host from humans to pest.

Any way you look at this, there needs to be some sort of trigger.


The problem, as other answers have allready pointed out, is that you need every instance of your virus to change in a very short timeframe.

Normal mutation obviously doesn't allow for this, so you need another factor that will cause the change. A global factor.

Introducing "Killer Cold Bacterium A (KCBA)": Under normal circumstances it acts like a mild cold. Some light headaches, mucus, and joint aches. Like a wart-infection it turns dormant, but never really vanishes. It also happens to be resistant to broadband antibiotics. Considering it's minor effects on the populace, Health Care Officials are reluctant to use more targeted Antibiotics treatments, because they fear the spread of further antibiotic immunity.

However, what we didn't notice was that KCBA also leads to increased absorption of nitrogen from the air. Under normal circumstances this doesn't cause harm, but after a melting glacier releases a significant amount of formerly bound nitrogen in the air, it causes mass-suffocation of anyone that is infected.


You could have a virus which has evolved to kill large numbers of social animals by only becoming dangerous when exposed to another chemical. So it infects a tribe and then they get killed off once they start sweating or coughing up something which makes the infections dangerous then a whole tribe can go down.

This could have evolved if the bacteria has a symbiotic relationship with a predator, something which might struggle to take down a pack of prey. The bacteria can then reinfect the Predator once it has come to eat the dead prey. The bacteria would have no negative effect on the predator though.

I don't know the biology behind that but as long as there is a benefit to it in an evolutionary sense it could evolve.

My take for your story

This happened years ago but the bacteria mutated to have a period of ninety years or so. It infected all of humanity but as a harmless bacteria. Now lifetimes are longer though, long enough for the bacteria to start producing the dangerous effects and no one suspects the elderly are the cause until too late.


Given the OP's question seems to imply this is a deliberate attack by a mad scientist or evil organization, then there should be a two stage process in play:

  1. Spread the infection. A genetically engineered virus or bacteria is released in such a fashion that it can reach and infect large numbers of people (for example, released at a major air hub or large metropolitan region). The organism infers the hosts, but there is no outward sign or anything amiss at this point.

  2. An artificial stimulus is released which triggers the infectious organize to change into its lethal form, causing mass die offs in a very short time-span, making it almost impossible for the authorities to develop any form of countermeasure.

The trigger would have to be something which can be quickly released and absorbed by the hosts in a short time frame, either a second organism, a chemical trigger or (very speculative) using some form of electromagnetic radiation as a signal to change.

The big problem wold be to find something which is both fast acting but not so ubiquitous that it might be stumbled upon by accident, alerting the authorities that something is amiss (mass die offs in an urban setting due to a relatively insignificant chemical accident might be a possible tip off). This sort of binary biological warfare agent seems unlikely in the extreme, but some sort of clever genetic engineering may make this avenue of approach possible.


Path 1.: Bioaccumulation or long lasting damage

What you're looking for is a disease that doesn't directly cause death. Think of something like Hepatitis B (not exactly what you want but close), or other diseases which can be carried unknowingly for the whole lifespan in some hosts.

Your disease would be harmless to all hosts but would come with long term damage.

Example: liver damage

All you really need for that is a harmful waste product that bio-accumulates in liver over months or years. I'll continue with liver example, but remember there are other parts of body that can get slowly damaged and it takes a while before you notice and then you die - brain, heart, CNS, kidneys, digestive tract...

Still, you - as a disease - would probably get found. There are people with liver issues who have their liver inspected, there are people getting liver operations etc. Doctors would notice something amiss. If goverments acted quickly on their advice and checked the healthy population livers, humanity might survive.

However the casualties would be pretty impressive still. Just imagine, 90% population with failing livers. Yellow skin and eyes, people vomiting and shitting themselves. Quite interesting from storytelling perspective.

Other examples:

CNS damage

You can make acetate or other organic solvent disease's waste product. These solvents slowly damage CNS. But people would probably notice the smelly breath and the dizziness.

Serotonin imbalance

Make it slowly disrupt serotonin levels (eg. again, by producing serotonin as a waste product). While serotonin imbalance doesn't kill you, it can make you kill yourself or someone else - what you'll be looking at are serious switches in mood, severe anxiety and sleeping disorders. Those people will look like zombies, but will be even more dangerous. If you want horror, this is a great path you can take with your story.

Great thing about this is that it's really hard for average person to admit to themselves that they have these switches in mood. You can expect mass craziness.

Path 2.: Competing diseases

Imagine another disease disperses along with the other one and, once they reach certain maximum levels in every host, they start competing in a harmful way. Imagine it being nuclear war between two diseases in your body. Not so realistic, but cool.

Bonus points for different symptoms in different strains.


As others have pointed out, you can't spontaneously apply a new mutation to every virus (or whatever) at once. If one population of the virus evolves into a lethal strain, the new strain needs to start infecting the human population from scratch. But suppose that the original strain, while otherwise harmless, weakens the immune system in a specific way that prevents it from being recognized as a hostile body. This is beneficial in of itself, since it allows the virus to remain in a host indefinitely, and not harmful to the host as long as no other illness exists that takes advantage of the weakness.

But suppose that after this original strain has infected the entire population, a new strain mutates in such a way that it becomes lethal after some incubation period. The lethal strain still needs to spread normally, but because everyone is already infected with the original strain, their immune systems have already been modified to ignore the new strain. If it's airborn, it'll spread extraordinarily quickly, even once it turns lethal and quarantine measures are put into effect. Presumably, there would be a few places that are isolated enough to keep the disease out, but most of the world would be helpless to stop it from infecting the majority of the population.


It's rarely ideal for a parasite to try to kill its host. Dead host means no more resources for the parasite, and that's no good. Therefore, most diseases limit the damage they do to their hosts. Most of the well-known exceptions are diseases that recently jumped between species, and they're not very successful -- Ebola, for example, doesn't usually spread very far before dying out, because it kills its hosts too quickly and too effectively.

BUT! That evolutionary calculation changes when another parasite is present. If there are two parasites present, competing for the same host resources, it turns out that it can easily become beneficial for each parasite to try to burn through those resources as quickly as possible, basically trying to take away those resources from the other parasite. "Burning through resources as quickly as possible" is generally not very good for the host. So, some parasites are harmless until their host gets another infection, at which point they switch over to a much more dangerous mode.

You could imagine an infectious disease doing something similar if it erroneously detected other copies of itself as a competing parasite. So if you get an infection once, you're fine, but if you get another one (say, a slightly mutated strain from a second source), it goes nuts and kills you. That way, the disease would be relatively harmless until it built up in the population, at which point you'd start getting double infections and deaths.


Realistic? And biological... Basically impossible, unless somehow a collection of viruses can exhibit intelligence...

Perhaps program your virus's evolution, somehow ensuring that certain traits would evolve at certain times, being able to customize what traits are evolved in.

Possibly some signalling could happen, somehow making your virus radio-sensitive or chemically or something, so you can broadcast a signal from your location, or the virus customizes humans to broadcast heartbeat signals towards other infected, giving data like, "some number infected, some number to go", etc.

Maybe this virus can infect anything, turning plants, and moulds and stuff into servers to direct the viruses that are inside humans.


The only ever scenario where something like this could be advantageous is in sparsely populated world of nomad humans, which have scientifically advanced medicine but don't ever communicate between them. Either because there are strong magnetic fields confusing radio waves, or just because they hate each other, for instance because of continuous splitting into religious sects which think everyone but them is straight-out working for the forces of evil.

Then the virus could choose to go undetected until it has infected the whole tribe, then kill off everyone to lay a massive amount of spores, which will infect the tribes taking the land or passing there.

This because the tribe could find a cure if it could see the illness in action, so the best strategy for the pathogen is to leave no witnesses. This works if tribes never communicate: every tribe coming in contact with the virus will be a dead tribe soon, so no knowledge accumulates towards a cure. Which is a pretty valid advantage for the virus.


As many others have pointed out there is no means of communication, and thus a Plague, Inc type switch is impossible.

However, lets consider a very skilled bad guy who wants to do this anyway. Can we come close?


Our pathogen must be a bacterium, of course it's made resistant to all antibiotics. Said bacterium likes to colonize the upper respiratory tract but causes no symptoms while doing so--but if the patient gets a cough for any other reason the disease goes along for the ride.

While the existence of the bacterium might be noted it will be ignored because it doesn't cause any symptoms. However, in time it is discovered that this "harmless" bacterium isn't so harmless after all, it just has a very slow incubation period.

Now, the bad guy is ready for this, he didn't think it was as harmless as everyone else and made a vaccine. The vaccine is legitimate and works, providing basically complete protection to anyone or anything it's administered to. (Said bug must infect something with a fairly short lifespan that is used for the testing. Perhaps a chicken.) However, if the vaccine is administered to someone already infected for a substantial period of time it's quite another story--the bacterium reacts by releasing a viral payload it has been carrying. This has an incubation period of weeks while it produces nothing but a mild cough, then it turns very lethal when the viral load gets high enough.

Thus the medical community provides the trigger, by means of something that to almost every test is safe and effective.


The key is to have the plague harmless and virulent. Preferably the body's immune system doesn't even recognize it.

Now the trick is to make it fatal once you get to a certain point.

If it was me, I'd use bacteria and have the bacteria pick up and store toxins that the body would normally handle in small amounts. The bacteria would release said toxin upon it's death.

Now if the bacteria used telomeres like human cells do, they would lose some every time the cell divided. Once the telomeres run out, the bacteria would die releasing the toxin.

Basically the bacteria would spread and spread but suddenly all around the same time, they would all die taking the host with them leaving mankind with no time to develop a cure



If you're looking for something that'll kill 100% of the population, I think you're out of luck without resorting to fiction.

The other answers rightfully point out that the organisms behind infectious diseases don't coordinate. But humans do. And infections can alter the effects of normal human behavior.

Make a virus that infects everyone with no symptoms but causes some major harm a while after a major human event happens.

Example: a virus that massively accelerates lung cancer in the presence of smog. This is definitely possible and similar situations exist in real life (for example, asbestos exposure dramatically raises the risk of cancer when exposed to other risks like smoking. It's not hard to imagine a virus that does the same thing).

Or maybe a strain of HSV-1 (which has infected the vast majority of the population) that drastically accelerates aging after a certain duration of infection (so that if you've been infected for a long time you just drop dead). Perhaps by shortening telomeres? This isn't really what you're looking for, but it is a human trigger in that now we live long enough to make this a problem.

It's also worth noting that most oncoviruses (cancer-causing viruses) probably wouldn't matter if life expectancy were shorter.

Edit: better example: a virus that has a powerful symbiotic effect on another infection. Pandemics are a relatively new phenomenon; pre-modern era, people were too separated for everyone to be infected. Now with globalization everyone can be exposed to relatively harmless infections. So give it a latency period that kicks into high gear later. Like a virus that waits 10 years and then kills you the next time you get the flu.

  • $\begingroup$ Welcome to WorldBuilding! If you have a moment please take the tour and visit the help center to learn more about the site. Have fun! $\endgroup$ – Secespitus Jul 5 '17 at 19:29

There are diseases where the initial infection is not lethal, at least with proper treatment. The HI virus comes to mind: the host is quite survivable, the virus can always spread by transferring bodily fluids*, and it's almost impossible to cure.

Another example would be the measles which can bring about some nasty complications years after the infection.

Herpes is a third example of a very persistent virus, though it's not nearly as deadly as required by Plague-inc.

* Is there any medication to make people less infectious?


Let's start simulating with a small and dense community, such as a single isolated town of 1000 people, to keep it simple. Every host frequently has contact to many others and the bacteria in infected hosts can form a network and stay in touch with all others. Now, after we have infected everyone with our harmless disease, we need to modify it to become lethal. I will consider multiple options:

Horizontal Gene Transfer

Bacteria have various methods to share DNA with another bacterium, most notably bacterial conjugation. This is for example used by some bacteria species to spread resistance genes. You start the trigger by implanting the gene in a small group of bacteria and let them share the genes with others. To modify the diseae in another infected host, only one modified bacterium would have to reach that host to start giving it the already present bacteria. This is a lot faster than letting the bacterium replicate from scratch. You could also include genes that somehow help to increase the chance and speed of the gene transfer.

Chemical Signals

Alternatively, the genes to produce the lethal toxin were present all the time, but the toxin wasn't produced because a trigger was missing. Most (bio)chemical reactions need a certain environment (eg temperature, pH value, presence of catalysators) to happen at a reasonable speed.

You activate the trigger by exposing some of the bacteria to substance A, which acts as a catalysator to produce toxin B as well as more A. The catalysator travels from bacterium to bacterium, signalling to start producing the toxin. You could also have A and B being the same substance.

Pulling the Trigger

Disclaimer: What I describe in this section is a stretch for reality-check and would require significant new inventions in the biochemical principles of nature.

Maybe we can use the complex nature of biochemical reactions to make our bacteria smart enough to "pull the trigger" themselves. Each bacteria could carry with it some signature information of every host it (or its predecessors or other bacteria it met) has ever had contact with. I'm not sure how to distinguish multiple hosts. Maybe we take a sample of the DNA (which should be slightly different in any host), maybe just a rough analysis of the chemical mix in the blood or whatever. It doesn't have to be exact, for we can never reach 100% accuracy.

These signature tokens should be shared amongst our entire network (but a bacterium will accept each token only once), so that at least some bacteria will have the entire set of the infected hosts available. All bacteria repeatedly "count" the tokens by putting a certain substance in the pool for each token. Once the concentration of that substance is high enough, i.e. the number of infected hosts roughly meets a certain predefined value, the bacterium starts the trigger.

Of course there is always the problem of mutations that let a bacterium start the trigger early, or getting the trigger from another source by accident. There can never be full certainty in biology. (And likewise I believe that it's absolutely impossible to kill off 100% of the world-wide human population with biohazards)


Could we lead the immune system of the body to help spread an even more sinister infection?

Consider it a two stage process with two different infections, assuming you have either a good distribution system or a well centralised patient zero:

  1. An infection that is highly contagious but does very little to the body, such that the host's immune system creates antibodies for it.
  2. Secondry virus would be very similar to the first so that immune systems uses those previous antibodies to fight it. But it is altered to use this unprepardness in a way to kill the host.

This is the way that I would try to develop it anyway. The first will pass by with very little fanfare maybe even being called a seasonal flu. But the second would kill, optimally I would like it to shut down the immune system so that even common infections will lead to deaths (evil laugh).


Attack mode - Spread by spore communication

I envision some type of flesh eating germ that normally aids in the breakdown of a dead body. The germ itself resides everywhere (like staph), but simply does not hurt people unless they are already dead. It remains fairly docile until the host's body dies, at which point it signals (through spores) to the other germs that they are in danger.

At this point, the spore signals cause other germs of the same type to move into rapid consumption and reproduction mode. Perhaps these spores would not be naturally airborn, but rather be transported by other means around the decaying body (signaling the germs to begin the decay process). Perhaps, this is due some type of stickiness of the spores that prevents them from leaving the body surface.

But the twist would be that there is either a mutation that causes the spores to be airborn, or some chemical introduced into the atmosphere which allows the spores to become airborn (maybe chemical warfare by a bad actor). This causes all these germs to enter the attack phase and release their spores. Suddenly every human dies within a short time period.


What you want is a "binary" disease. Each part is relatively harmless but when mixed together, they become dangerous and usually lethal.

The first part should be some sort of bacterium or virus that can be easily spread, such as an influenza virus or a new version of the hundreds of viruses that cause the common cold. Since there are always new strains of influenza viruses developing, this will not cause much concern. Nobody worries about being exposed to a new strain of the common cold or the flu. They just buy the local remedies to alleviate the symptoms until it goes away. Perhaps it is even included in a version of the flu vaccine. In this case, the virus would only be dormant.

When you have ascertained that enough people have been infected (perhaps by reading various CDC publications such as Morbidity and Mortality Weekly), you introduce the second portion, which need not be a virus or bacteria. This chemical needs to be easily absorbed through the skin and which can be easily added to something people use every day. If this chemical, which is harmless on its own, was some sort of additive that allowed a printer to use less ink, then your government would probably buy it and use it for enormous runs of something it prints and distributes to everyone, like tax forms or money.

When the chemical is absorbed through the skin, it triggers a mutation in the original virus. The mutation is lethal but has a dormancy period that allows the spread of the deadly mutated virus far and wide. Your mutation could be similar to some virus that does not have a cure, such as Ebola.

When people start dying, the deaths will be widespread and rapid. Even if they discover that your additive was the culprit, you would not be suspected of planning anything like this. It would be one of those mysterious anomalies.


protected by Community Jul 28 '17 at 7:21

Thank you for your interest in this question. Because it has attracted low-quality or spam answers that had to be removed, posting an answer now requires 10 reputation on this site (the association bonus does not count).

Would you like to answer one of these unanswered questions instead?

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