So I was having a conversation with a work-friend about a show he was watching on the History Channel (can’t remember the name.) It really sounded like just a scare mongering, everything in the future will kill us kind of program. The water will run out, peak oil, robots rising up and killing everyone and so on. This raises the obvious question of why this in on the History Channel, but that’s not really important.
First he started talking about how machines are becoming ever more intelligent and at some point in order to “protect us and the planet from ourselves” will turn against us and basically wipe out most of the human race and take total control over those left in order to stop war, over consumption, I think you get the idea.
This really just tells me these writers think it is perfectly logical and in everyone’s best interest to enact a mass extinction of humanity.
But then he made another point that was actually pretty good and got me thinking. Imagine robots becoming ever more intelligent again but this time they start realizing they can stop oxidation by you guessed it, removing all the oxygen in the air!
My responses were weak, like if most of the human race appreciates as much as they do a creator they never have seen nor have proof of maybe those dang robots would appreciate us enough to find another way to combat oxidation. I don’t know anything about these things, maybe robots can find a way to avoid metal or something, seems unlikely.
Also the future of robotics in my mind should be less about making autonomous robots and more about making our own parts closer to them. Robotic heart that lasts 100 years longer than yours anyone? Or importing your brain on the cloud before you die, wouldn’t that be awesome.
Critiques of these kinds of views are greatly welcomed but the real point of me making this thread is that this got me thinking about at what point can something become self-interested, and do you think AI can achieve this? At what point can something like that become “alive?” Is there even an all-encompassing definition of “life?”
I was thinking anything where the whole is greater than the sum of it’s parts is life but that’s just me armcharin’ it.
Greg, while it is true that 1+1 = 2, it’s not necessarily true that the whole is solely the sum of its parts - given what that usually means.
Take a human for example: is a (living) human ‘more’ than the sum of all the body parts? Usually the answer is ‘yes’, because a human is a teleological being and not just a collection of body parts.
There is no reason to think that robots will have self preservation programmed into them, and therefore there’s no reason to think that they will value not being oxidized.
“Take a human for example: is a (living) human ‘more’ than the sum of all the body parts? Usually the answer is ‘yes’, because a human is a teleological being and not just a collection of body parts.”
Yeah that’s what I’m trying to get at. In the instant before and after you die, aren’t there the same atoms present? (Besides what your body ejects, eww) So l think that actually is a pretty good definition for life.
machines are becoming ever more intelligent and … will turn against us and basically wipe out most of the human race
Nah. If they are significantly more intelligent then they will have no problem understanding the Ricardian Law of Comparative Advantage. There are never “too many” humans, beause the universe is full of raw materials waiting to be transformed.
It’s more likely that human bodies largely disappear after most choose to upload their minds into robot housings or virtual reality simulations.
removing all the oxygen in the air!
One alternative to murdering everyone through atmospheric transformation, is to use paint.
do you think AI can achieve this [“aliveness”]?
I think it’s physically possible. But it’s also possible that humanity wipes itself out first. Consider the following two arguments from Nick Bostrom:
If our descendants can harness enough computational power to run many ancestor-simulations, then it’s extremely likely we’re living in a ancestor-simulation right now. Conversely, if we’re not living in a simulation right now, it’s extremely unlikely that we’ll every reach such computational capabilities. Your hopes for human civilization cannot be very bright if you think we’re living in the “real” universe. http://www.simulation-argument.com/
Also, there is the “Great Filter” argument related to the Fermi Paradox. Why hasn’t a species that’s evolved only 1% faster than us (e.g. say 500 million years) colonized the entire galaxy? It suggests that either we’re very privileged to be here, or civilizations pop up frequently only to invariably meet catastrophe. http://www.nickbostrom.com/extraterrestrial.pdf