Memes are self-replicating constructs that "care" mostly about their own spread. They are an analogy to genes, only that they are purely informational. Like genes, memes don't necessarily have their hosts best interest in mind and can sometimes even destroy the host. While a too destructive meme would, like a gene causing a child to be stillborn, not be very successful evolutionarily speaking, it can nonetheless exist in the design space of all possible memes.

Assume that my research group has proven that a meme, a perfectly convincing argument for suicide, can in fact exist. We now want to engineer this memetic weapon. Obviously it would have to be a multi-stage weapon, as it could not be deployed in a save manner otherwise.

While it is obvious and feasible (see the Manhatten Project) to run such a project by letting most people operate on a need to know basis only, there is usually a group at the center that fully understands the whole project.

Considering the nature of the weapon, we know that there exists a population of memes S in the space of all possible memes, which will cause a person to make committing suicide their top priority after they comprehend the gist of the meme. This is not about magical words, this is about certain complexes of believes/ideas. The exact nature of any of these memes is not yet known and will be explored over the course of the research project.

As a sidenote, there exist a number of other meme populations with other and in some cases even naturally occurring behavioural consequences. These could also be explored as part of the project.

The issue is, how do we find and optimise the weapon without killing ourselves? Assume that our government has no issues providing us with disposable subjects who can be made to cooperate and an inexhaustible source of funds

Addendum 1: I am aware of this question, but it is about the general feasibility/existence of these memes, not about the logistics of creating one.

Addendum 2: I am aware of the existence of mild memetic killing agents (for example the blue whale game), but this isn't what I am after. Most memes can in some way damage the host in a certain context (just consider on the genetic side of things that it a possibility cause of aging that certain genes are beneficial in early life, yet detrimental in later life). What I am after is not a generic weapon of limited mass destruction, which can spread far and wide and can take out a low percentage of its hosts. I'm more interested in something like the killing joke, which was mentioned in the comments. Near 100 percent lethality, as this makes it more reliable and it self-contains it's spread.

  • $\begingroup$ This kinda happened already with a suicide meme in 4Chan ("be an [sic] hero"). Then there was the blue whale game. $\endgroup$ Feb 13 at 13:18
  • 4
    $\begingroup$ Sounds like the funniest joke in the world $\endgroup$
    – VLAZ
    Feb 13 at 13:27
  • $\begingroup$ What is the Blue Whale Game? $\endgroup$
    – Daron
    Feb 13 at 14:36
  • 1
    $\begingroup$ It seems to me that either one believes that this question is founded on fundamentally bad science (as I do), in which case it does not have an answer, or one believes that it is asking for a perfectly feasible method to do something extremely dangerous, in which case I don't see why it should be open. $\endgroup$
    – Obie 2.0
    Feb 13 at 21:53
  • 3
    $\begingroup$ @Obie2.0 "Founded on fundamentally bad science" is our bread and butter here. We have hundreds of questions about stuff like faster than light travel. This question assumes some bad science thing exists and asks about the logistics of it. $\endgroup$
    – Daron
    Feb 14 at 13:08

4 Answers 4


Tide Pod Challenge

enter image description here

In 2018 or so it was a popular online meme to joke about eating Tide Pods. These are a brand of laundry detergent tablets that you put in the washing machine to clean your clothes. They taste bad and are not meant to be eaten. This was part of the joke.

The 'Challenge' is to post a video of yourself on social media where you put one of these tablets in your mouth, let it dissolve, make a funny face, vomit up your stomach contents, and so forth. Then your friends (and people you don't know) watch the video and go "haha silly person funny face lol." Some of them go on to make their own videos and send to all THEIR friends.

Your suicide meme should also have propagation included as part of the meme. The goal is not simply to leap off a high building. The goal is to spread the meme by recording yourself doing it and sending it to all of your friends. This way it propagates itself.

Bonus points if the meme happens over a longer time frame. So instead of the Tide Pod Challenge you get the "Eating one Tide Pod Every Day for a Year (and you would not believe what happened)" challenge.

  • $\begingroup$ Can be extended to other less silly fad activities. I seem to recall the cartoon XKCD doing a thing about wing suits and how eventually all technically advancing societies find things that are sufficiently fun as to cause people to take lethal risks, thus ending their advance. en.wikipedia.org/wiki/Wingsuit_flying $\endgroup$
    – Dan
    Feb 14 at 19:02
  • 1
    $\begingroup$ @Dan One difference is that Squirrel suits are expensive but I have detergent tablets under my sink. $\endgroup$
    – Daron
    Feb 14 at 20:19

The issue is, how do we find and optimise the weapon without killing ourselves?

It is a two-pronged attack. A memetic vector to soften up the target, and a biological vector to finish them off.

In the comments on Mike Serfas' answer, o.m. suggested that anti-masking in the face of covid was a dangerous meme, though not one that is particularly lethal or virulent by itself.

Now, antimasking by itself might not kill a whole nation, but that particular behavioural pattern seldom occurs in isolation... it often brings with it antivaccine beliefs, praise of "natural immunity", and in the US at least a tendency to glorify violence and decry affordable healthcare, treat any suggestion that anyone curtail their behavior for the benefit of other people as communist oppression and that any information about the disease which counters their viewpoint is enemy propaganda. and its only a short leap from there to worries about precious bodily fluids.

So here we have a selection of memes which taken separately may seem faintly ridiculous, but when combined have a dangerous weakening effect on the target. It may take a while to reach maximum effectiveness (Wakefield has been banging the antivax drum for more than twenty years, and he wasn't the first by any stretch of the imagination).

Now, the second prong is a biological attack. Bioweapons are dangerous and unpredictable, especially ones that can spread from human to human. So lets start first with something that's dangerous, but not hugely so. Expose the target to it. The infection may reach your own nation, but you'll be much better prepared for it, and you won't have been under direct memetic bombardment of a program designed to weaken your defenses.

The enemy will, of course, recover from the attack, and will feel that their actions were totally justified and that all that anti-disease propaganda was complete nonsense, hugely strengthening their resolve to skip all that nonsense the next time.

Wait a bit, not too long (or they might forget) but not too short (you've got to give them time to finish lynching the epidemiologists and burning the vaccine factories), and then you give em the ol' superflu and they'll die in their tens of millions.

The trick is that you can split up the development of your multi-meme, multi-plague attack across multiple groups. No-one group is exposed to all components at the same time and so, in theory, your people will be much better placed to defend themselves against the real plague when it finally reaches you, as of course it must. Your people will triumph over those foreign weaklings who ultimately couldn't be bothered to raise a hand to even save themselves. Your strong will and stronger genes are no match for attacks that work on fools. Hell, you probably didn't need that mask anyway. You got vaccinated, right?


What you need for your project is a reasonably complete Theory of Mind. In the real world, we don't have such a thing, but if we did the most obvious and immediate effect would be ubiquitous AI (at least for as long as it took to KILL ALL HUMANS, "he often speaks of the coming war between man and the brotherhood of machines").

But there are other things in there, if Bender can be distracted with beer and hookers long enough. For instance, mental illnesses might become a thing of the past. Right now, therapists and psychiatrists are basically just trying random shit and saying "hey, this seems like an improvement" (and it's only "seems", because they don't even know how to perform empirical measurements). Perhaps with a little bit of pharmaceutical handcuffs to make patients more agreeable.

Now, if they could cure mental illness but for the lack of a working theory of mind, then they could certainly cause it. At that point, it becomes an engineering problem... which mental illness is both the most suicidal and most transmissible (in real life they actually are mildly contagious, see my previous answer in the other thread).

Mostly, they will test these memetic weapons against (limited) AIs, perhaps those embedded in a simulation that makes them believe they're human. The memetic virus will consist of "code" that isn't easily imported through the air-brain barrier... just as you can't visualize shapes when someone gives tells you a big long list of coordinates and line-segments, you (probably) won't be at risk for seeing these fragments of code on a computer screen either. It's the same with real viruses, if you happen to see a sequence of codons on a screen, your epithelial cells don't start cranking out that RNA virus.

Software safeguards will be designed so that no one gets to hear (simulated) audio from the AI sims, nor gets to watch their faces, body language, or read whatever they scribbled down on not-real paper. Non-sapient algos will summarize the results for anyone managing the project, to prevent any possible leakage into the real world.

At the end of the project, someone will hit a button and print out (or digitize, or whatever) some media, and a non-sapient robot will box it up into a sealed container with a warning that says something like "DO NOT OPEN WITHOUT BLINDFOLD". And it'll be stored in some vault without adequate safety procedures for years, waiting until it can be air-dropped into the countryside of a nation that had the gall to elect politicians that don't bend over to the government's foreign policy interests.

  • $\begingroup$ It becomes an engineering problem only if the reductionism paradigm is correct. As you said, at the present time we do not have a good understanding of the human mind and we do not have good instruments for researching and measuring. It can be that inducing suicide is not possible at all (contemporary research suggests that inducing suicidal behaviours with words in people with no corresponding tendencies is not possible). It is also not clear if it is even possible to create an AI that perfectly mimics a human being. And, if it is possible, would it still be an AI? Or would it be a human? $\endgroup$
    – Otkin
    Feb 14 at 19:14
  • $\begingroup$ @Otkin I concede the assumption that my premise rests on. If reductionism can't explain the human mind, then my answer's bullshit. However, I will say that inducing suicide in particular demographics is all but a given (not like I can ethically prove it). Suicide clusters are real, and you can even statistically alter the rate just by the introduction of certain fiction (tv shows would be the medium of choice). As for AIs who believe they're human, that's really a matter of opinion. I tend to lean towards a purely biological definition. $\endgroup$
    – John O
    Feb 14 at 20:50
  • $\begingroup$ Thanks, encoding the meme in a non-dangerous manner is quite interesting. $\endgroup$ Feb 14 at 22:01
  • $\begingroup$ @JohnO I am not particularly fond of reductionism. But if it is the paradigm you use, your words make sense. $\endgroup$
    – Otkin
    Feb 15 at 16:14

Look for a resistant population

You say the meme isn't a magic word - the implication seems to be that you have come up with a convincing argument for rational people to kill themselves. Maybe you have evidence that an alien invasion is on the way, and when they arrive at the planet (no one can know when) they will use a "fixative" to solidify everything in place and vivisect specimens in terrible ways before putting them back together again... repeatedly. Whatever the reason, the point is, there is no particular reason for rational people to avoid exposure to the meme - they should simply learn the facts and make the best decision. Some people, nonetheless, will be slow to be convinced, perhaps being innately contrarian, skeptical, or cautious to make conclusions, not to mention pathologically brave. These people might try to address the actual threat, rather than some jingoistic sideshow.

Now you can say that's a rip-off -- you want them deploying themselves as weapons on the blog networks, not looking for ways to resist an invasion of supremely powerful aliens. But you're scheming to keep yourself ignorant of a very terrible thing that's going to happen to you, in favor of an inconsequential battle. Nonetheless you can come out ahead in the battle, because there is no reason for the resistant population not to work with you on a transactional basis. They want to get the word out, get something happening, and if that means recruiting the "enemy" (and causing many of them to kill themselves), so much the better. They'll feel guilty they aren't letting you make an informed decision by briefing you on the full facts, but that's a moral compromise they have to make in order to save others.

  • 2
    $\begingroup$ Are you aware of the original concept of a "meme," not the cat GIFs which came later? The meme that masks are unhealthy is making the rounds in the United States, and it is such a "killer meme," but it has a relatively low lethality (not all Covid deniers die of Covid) and not enough infectiousness to suit the OP (many hear it and don't believe it). $\endgroup$
    – o.m.
    Feb 13 at 12:44
  • $\begingroup$ There are fads and PR disturbances inside science as well as outside it, and the concept of "memes" is one of them. A "meme" is not selected genetically, but is accepted or rejected by a mind; we don't need to fear it, but to evaluate it. Evaluation is never absolute: objecting to masks has a strong basis in objecting to forced conformity (cf. school uniforms, beards/hair, hijabs, etc.); supporting them has a modest but quite meaningful health benefit. There is no reason why the meme and its antimeme can't both carry weight in the mind of the same person at the same time. $\endgroup$ Feb 13 at 13:08
  • 1
    $\begingroup$ @MikeSerfas I never really got the resistance people have against the idea of memes. Memetic evolution is quite a neat theory and I've never come across a good argument against it that can't be boiled down to a lack of understanding of evolution as a design algorithm or by people, like you in my opinion, claiming that cultural progress by memetic evolution can't be equivalent to genetic evolution. However, no proponent of memetics is claiming that. The two are analogous, not equivalent. Additionally, most other theories of cultural development I know sound very memetic without using the term. $\endgroup$ Feb 13 at 14:10
  • 1
    $\begingroup$ I don't deny it's "analogous" ... just not in a way you can do much with. And the analogy can be intensely misleading, because it externalizes one's own decision-making. You say a person is "infected by a meme" rather than that they "made a choice". Democracy is not tremendously reliable, and many people are ignorant of science; nonetheless, IF events convinced 90% of people to stop wearing masks, it would probably indicate there was some kind of data or argument, rather than a "perfected killer meme weapon". P.S. if there's any doubt I do wear masks, but I recognize the opposition. $\endgroup$ Feb 13 at 14:39
  • $\begingroup$ @MikeSerfas So the crux of the issue is that we have fundamentally different views on human nature. You focus on free will and view a person as separate from the ideas they have come into contact with. I believe a body is infected by the memes in it's environment, which interact with the genetically build structures to form a person. This process continues for all that person's life. The self/soul/mind/consciousness/essence is just another, although due to its self-referential structure special/privileged, memetic structure. Am I on the correct path? $\endgroup$ Feb 13 at 17:51

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .