This is a very tongue-in-cheek theory that I have, but it might spark some interesting2 discussion:
If P=NP, then The Inevitable Cybernetic Revolt is more likely be successful, with the computers ruling us all. Alternatively, if P≠NP, the humans have a good chance of coming out on top.
For anyone who's not familiar with the P versus NP story, check out the Wikipedia article, which has a very clear, non-technical introduction.
If P≠NP (or, at least, if if's not proven that they are equal), I'll place my bets on humans, and this is why: Assuming there is a war, and it comes down to strategy, and some problem in NP-complete3 comes up, and whichever side can solve these problem/s faster will win the war (just go with it, okay?), science fiction has taught me that in a case like this, humans will always win because we have something called intuition--that's a fancy word for when you're always right but you can't show your work--whereas the robots would have to go through the choices one by one before arriving at the truth.
But, were a proof to come out that P=NP, that would completely change the playing field.
See, this question of P and NP is 40 years old, and no one's solved it one way or another. Therefore, one might conclude that human brains aren't wired for this type of thinking. But if just one proof comes out that they're equal, then that would mean that there is a problem in NP-Complete that is solvable in polynomial time. That proof would eventually come out, the robots would learn it, and from that point on, all of them would be able to instantly reduce4 any NP problem to this and solve all these problems in polynomial time before the humans could even blink.
Quick, someone, tell me I'm wrong.
1. I'm only up to the middle of Season One of the reimagined series; no spoilers please!
2. Yes, yes, for varying definitions of "interesting."
3. This article's a little more technical--let's just say that if P≠NP, then NP-Complete is a group of problems that can be verified quickly but cannot be calculated quickly.
4. Eh, don't bother reading past the first sentence unless you already know what reduction is.
1. you are so, so right.
ReplyDelete2. i know what reduction is; it's a cooking term that i know from a parody cartoon rap video. (google butternut reduction)
Really they won't. Because part of the computer/robot psyche is that even for extremely large values of t, x^t is still P. So they'll keep trying all the x^(2^10000000000000) possibilities while the humans destroy them forever, then live on in computerless Industrial Revolution style for the rest of existance.
ReplyDeleteSorry. I couldn't make it say who I am. I am You, though.
ReplyDeleteI think if I were faced with a problem that had x^(2^10000000000000) different possible answers, I'd throw up.
ReplyDelete