That’s when a lightning bolt fuses the power cord to the outlet. Meanwhile, a second lightning bolt strikes you dead.
I can’t see a downside to taking both boxes. The hypothetical is explicit in saying that you have no control over whether there is or isn’t a million bucks in the opaque box. For those that wouldn’t take both boxes, why not? You don’t lose your thousand. If you don’t take the opaque box along with the thousand box, you don’t get a million. If you do take both, you just might get the extra million. Again, what’s the downside? Let me emphasize again that you get to keep your thousand dollars even if you take both boxes. It doesn’t cost you anything to take both and it might cost you one million dollars to leave the second box.
I don’t really see that it says anything directly about free will but it certainly illustrates the human predilection for superstition. How quickly we get the nascent Cult of the Great All-knowing Supercomputer from the explicit condition that you have no control over what the computer decides. Fascinating stuff.
I said that I would take one, but I somehow assumed that if you took two you got only the contents of the second. Rereading the OP I realize there is no penalty for choosing two so you might as well. A much better dilemma would be to make you choose between the first or second box. I would try to make my choice random (e.g. flip a coin).
Yeah. I don’t see the dilemma?
The sorta dilemma is that if you take two boxes, the infallible computer knew you’d take two boxes, and so the second box is guaranteed to be empty. Zonk!
The “trick” such as it is, is to persuade the computer you’re the type who’d only take one. Then actually take two. The dilemma is how to so persuade the computer. Which is doubly hard when the darn thing is infallible.
Overall I vote this is a dumb problem / thought experiment, or I’m too dumb to understand its subtleties. Or something got garbled in the problem statement.
I know that reference! Answer, by Fredric Brown. He was good with short-short stories like that.
Part it out. The memory alone is probably worth more than a million dollars by itself.
There really is the question of what is the rational thing to do in this hypothetical situation, and, moreover, to give a formal proof that it is the correct strategy (including formalizing non-paradoxically what the computer is supposed to be doing), since, as has been pointed out, there are plausible arguments for either (A) taking one box or (B) taking two boxes. But they cannot both be the right move!?
The real question here is, what would Captain Kirk do to make the computer self-destruct?
I’m going to join the others saying they don’t see the dilemma. There’s no real downside to your actions at the moment. If the computer thinks I’m greedy and I grab both boxes, I get $1K. If the computer thinks I’m not greedy and I grab both boxes, I get $101K. Whether I actually grab both boxes doesn’t really enter into it, and I don’t see why you’d try to please the computer by grabbing only one box. It only depends on whether the computer thinks I’m greedy, any payoff in my actions would depend on me fooling it in the past somehow.
You would do it to get more money, not to please the computer. It’s just a computer.
I tried finding a site that hosts a (good) Rock-Paper-Scissors bot, but it seems to be down at the moment.
Add a box.
Take both.
The supercomputer’s predictions are basically a coin flip that happens after you open the box, at least from your perspective.
So you’re presented with these two options:
- Take the $1000 box, and leave the coin flip.
- Take the $1000 box, and take the box and do the coin flip later.
Why wouldn’t you take both boxes? You’re effectively trying to second-guess a coin flip if you do anything else.
Honestly my initial take on entering the room is “whose money is this? They shouldn’t just be leaving a thousand bucks in an open box? Someone might just take it.” The computer says it’s mine if I want it and gives me the set up. “Meh. This is what you are programmed to say but I don’t have any reason to trust you. I could get arrested tomorrow for stealing the money and my defense is a computer told me it was okay? No thank you .”
I wouldn’t play. I close the box. Close the room. Walk away.
No. The option of taking the $1000 and leaving the mystery box was never presented as an option (and really, it’s the only illogical choice). These are the 2 presented options:
- Take the $1000 box and take the mystery box and do the coin flip later.
- Leave the $1000 box and do the coin flip.
If $1000 isn’t enough to move the needle, the coin flip makes perfect sense. I understand the rationale underlying both options. What I don’t understand is the adamance that there is only one reasonable choice.
Economy of scale I think makes a difference here. Think about the scenario like this:
- The open box has a dollar in it. The mystery box has either nothing or $1000 in it.
- The open box has a million in it and the mystery box is either empty or has a billion in it. (This scenario would make me a two-boxer)
I have a very hard time understanding why nearly half are saying they would take the one box.
What is the downside of taking both? Too hard to carry? You are so sure that the computer is infallible because it has been right 1000 times in a row that why bother to take it?
The possible outcomes if one plays are: one box with a thousand; one box with a thousand and a cool opaque box to show off as a souvenir; two boxes one with the thousand and one that can be thought of as a free lottery ticket - unlikely to win but nonzero and it’s free. Why not take the free lottery ticket?
If you take it and the computer predicted you would you do not get the million dollars. The only way to get the million is for the computer to predict you will take just one box, and since the computer is batting 1,000 for 1,000, it seems safest to just trust that the computer will predict correctly.
This right here is why so many people buy multiple lottery tickets.