If over 50% of people choose the Blue button, everyone lives.

If less than 50% choose Blue, only the people who choose Red live. Those that choose Blue die.

It's an interesting social experiment and dilemma.

I'd like think I would choose Blue because I would want all to be saved. It's not worth living if I feel I have the blood of others on my hands. But I also like living and we are living in an increasingly selfish and low trust society… but would you really want to live in that society?

Everyone should just choose blue, then everyone would live.

Blue is the only ethical choice. Advocating for red is the only utilitarian choice.

I think I'd have a hard time living with myself if I choose the red button.

Howdy, it is me! said:

---

Great example of how we all see and think about things differently. I'm sitting here thinking, "Will everyone realize that if we all pressed red we would all live?". But I'd press blue because I don't think 100% of people would realize that (as demonstrated by this thread).

---

Agree.

the premise is that you don't get to collaborate before hand.
if 100% of people choose the same color, then everyone lives.
if only one person chooses red, everyone lives.
If only one person chooses blue, that person dies.

so everyone should choose blue, to protect that one person.

Ultimately, this is a test of your expectations of other people.

If you assume that most people will act in the common interest (press the blue button), then it's safe for you to do so yourself.

But if you assume that most people will act in self-interest at the expense of other people (press the red button), then you're faced with a dilemma of whether to join them and kill other people in order to save yourself.

The collective worst-case scenario (maximum death) is achieved when 50.1% of the population chooses the red button and 49.9% chooses the blue button. The best-case scenario (nobody dies) can be achieved by 100% red or majority-blue.

I'm sure there's a political analogy to be made with this.

Just for fun, I posed this scenario to ChatGPT:

Quote:

---

This is a classic coordination problem with a twist of risk and trust. It blends ideas from game theory (like the Prisoner's Dilemma) and threshold coordination problems.
First, strip it down to incentives

• Red is individually "safe." No matter what others do, Red guarantees survival if Blue doesn't reach a majority.
• Blue is collectively optimalbut fragile. It only works if enough other people also trust the system and choose it.

So the tension is:

• Self-preservation under uncertainty (Red)
• Collective trust and cooperation (Blue)

What makes this problem tricky
Unlike a simple Prisoner's Dilemma, this isn't symmetric:

• If everyone picks Red everyone lives (stable but pessimistic equilibrium).
• If a majority picks Blue everyone lives (optimistic equilibrium).
• If trust fails just slightly catastrophe for Blue choosers.

So there are two "good" equilibria:
[ol]

Universal caution (all Red)

Majority trust (Blue wins)

[/ol]
But only one of them (all Red) is risk-dominantit doesn't depend on trusting others.

What I would choose
If I'm being strictly rational under uncertainty: Red.
Not because it's morally betterbut because:

• You cannot communicate.
• You cannot verify others' reasoning.
• A small failure in coordination leads to massive asymmetric harm.

Choosing Blue requires confidence that a majority of humanity will:
[ol]

Understand the game correctly

Believe others will cooperate

Act on that belief

[/ol]
That's a very high bar.
So Red is the robust strategy; Blue is the high-risk, high-reward gamble.

What different thinkers might do
Social contract & cooperation thinkers

• Jean-Jacques Rousseau
  Likely Blue. He believed humans are naturally cooperative and that collective good can prevail if people act as a unified body.
• Thomas Hobbes
  Almost certainly Red. His view of humans in a "state of nature" is distrustfulpeople act in self-preservation first.
• John Locke
  Probably leans Blue, but cautiously. He believed in rational cooperation, but also in individual rightsso this one's a toss-up.

Moral philosophers

• Immanuel Kant
  Likely Blue. His categorical imperative pushes toward actions that could be universalized"act as if everyone will do the same."
• John Stuart Mill
  Strongly Blue. As a utilitarian, maximizing total well-being (everyone lives) dominates.

Political and military leaders

• Winston Churchill
  Probably Red. He was willing to take risks, but not blind onesthis scenario lacks control and coordination.
• Napoleon Bonaparte
  Likely Red. He preferred strategies where outcomes depend on his own decision, not mass psychology.
• Mahatma Gandhi
  Likely Blue. His philosophy depends heavily on trust in collective moral action.

Modern game theory perspective
A theorist like John Nash would point out:

• Both (All Red) and (Majority Blue) are equilibria.
• But All Red is safer because it doesn't require coordination.
• Without communication, systems tend toward risk-dominant equilibria, not payoff-dominant ones.

The real crux: What do you believe about humanity?
This isn't just a math problemit's a belief test:

• If you think most people will play it safe Red is correct.
• If you think most people will trust others to cooperate Blue is correct.
• If you think people are split Red is overwhelmingly safer.

A subtle twist
There's an irony here:

• If everyone reasons like a cautious game theorist, Red wins (and everyone lives).
• If everyone reasons like an idealistic moralist, Blue wins (and everyone lives).
• The disaster only happens when humanity is mixed and uncertain.

So the worst outcome isn't selfishnessit's lack of shared expectations.

If you want to push this further, we can explore how adding even tiny communication (like polls or signals) radically changes the outcomeit becomes much more likely that Blue succeeds.

---

An important point of the scenario is that the players can't communicate about their choices.

If a poll is taken before everyone presses the buttons, and 80% vote Blue, then the Blues can feel secure in their choice, and maybe a few Reds will flip in order to not seem selfish.

If 80% vote Red, then the Blues will know that they either have to switch to Red in order to live, or be prepared to die for their collective-good idealism.

It's when it's 50/50 that the scenario gets the most dilemma-y.

if your toddler gets to press a button and their favorite color is blue, then you better pick blue to protect them.



Red only protects you. Blue possibly protects everyone.

swimmerbabe11 said:

---

if your toddler gets to press a button and their favorite color is blue, then you better pick blue to protect them.



Red only protects you. Blue possibly protects everyone.

---

I think the setup works best when everyone understands the rules (or has a guardian who does) and is making a deliberate choice (without communication, as Bonfire noted). That's what makes it an interesting reflection on human nature and what we think of human nature.

Once you have kids or toddlers randomly choosing based on something like favorite color, it completely changes the problem into more of a personal ethical question (whether i protect myself at the expense of millions and millions of babies/toddlers/kids). In that case I agree the answer is blue.

Howdy, it is me! said:

---

Great example of how we all see and think about things differently. I'm sitting here thinking, "Will everyone realize that if we all pressed red we would all live?". But I'd press blue because I don't think 100% of people would realize that (as demonstrated by this thread).

---

I think this is exactly right.

I do understand the logic in selecting red, but in a scenario where we do not all get together ahead of time to select the same color, there will be a split. If this were actually put to everyone on the planet, I think there would absolutely be billions of people in each color group.

Imagine the unlikely scenario where 50.001% vote red and 49.999% vote blue. If I was in the red camp in this scenario, I don't see how the guilt of my decision to not try to save 4 billion people doesn't completely consume me.

"The single raindrop never feels responsible for the flood."

This is why an all knowing God makes sense and this human experiment is moot.

In the original prisoner's dilemma there was a potential negative outcome for doing the equivalent of pushing the red button that is absent in this scenario. Sort of changes the way we look at it. Lacking the potential negative outcome it seems like there is more incentive to push the red button.

I'd push blue, but I have no real confidence that >50% of people would do the same.

What kind of death occurs for us if blue doesn't make it?

In all reality, the winner here if this was a REAL thing would be red. When it comes to self preservation, human nature wins.

The version of this I saw phrased this as "everyone on Earth"

That means that people that think like me are a tiny minority. The outcome will be determined in places where almost everybody who is participating in X polls would likely experience extreme culture shock by going there. I wouldn't even feel comfortable attempting a prediction. The result might be like 90/10 either way.

50% is also arbitrary. Like would many people pick blue if the arbitrary safe zone was 95% blue? The only real argument I see for picking blue is to save those who aren't smart enough to pick red. Its not a great argument. I understand the ramifications. I have a three year old daughter. Should I pick blue to save her if she pocks wrong? What if she picks red and me and my wife die in an 80/20 landslide leaving her orphaned? Its just as likely of an outcome.