[Creecher]

Spoiler

This video brought some questions to mind about Androids. With technology becoming more and more advanced, it may be possible that even our creations will be capable of human emotion and "thought".

While we are still a while away from having androids like the one in the video, we can still have a conversation with an autonomous computer in this day and age.


What do you think?

This post has been edited by modi123_1: 12 February 2013 - 07:57 AM
Reason for edit:: added spoiler tags

[creativecoding]

I wouldn't say we can have *real* conversations with computers. The closest we have is something like cleverbot, that basis it's responses after what it's been told in the past. I don't think true conversations with computers will come until they have the ability to think for themselves, which, unfortunately is also based off of things they will have learned. But that's not exactly the focus here.

I would say it's definitely possible to create humanoid robots capable of human emotions and thought, just not anytime soon. The human brain is essentially a type of computer, processing information and responding to it.

This raises some interesting questions though. Would these robots have rights? Why and why not? At what point do they stop being robots, and start being human? Scientists have been able to put brains (lamprey brains - not human) inside of robots before, and have them operate. What if that were to happen? Would they have rights then? Would robots be responsible for their own actions?

Honestly, I think before any sort of robot like that is achieved (emotions, thought), we'll be seeing "smart" robots. One that are on a real basic level of protect and serve. And I think that'll truly be the best, because one people start focusing on making them more and more human, eventually you'll get to the point where you get robots wanting to do their own thing, have paying jobs, hold places in government - they'll want their own rights and we'd be unsure whether or not they technically deserve/should have rights. There would probably be some form of segregation for awhile. Robotism? Maybe. People being accused of being robots and killed because of it? I wouldn't be surprised (however I would be surprised at how realistic those robots at that time are).

It's certainly and interesting field and I think the changes it'll bring will be tremendous.

[lordofduct]

creativecoding, on 12 February 2013 - 10:50 AM, said:

I wouldn't say we can have *real* conversations with computers. The closest we have is something like cleverbot, that basis it's responses after what it's been told in the past. I don't think true conversations with computers will come until they have the ability to think for themselves, which, unfortunately is also based off of things they will have learned. But that's not exactly the focus here.

Your logic here is kind of backwards. What is a "real conversation"? What is the definition of it? It sounds to me that your definition for it is a conversation with something that can think for itself. So of course you're not going to have one with a computer until it can think for itself. You defined it as such.

And what is think for itself? What's the definition of that?

My computer thinks for itself. I give it inputs, it thinks, it gives me outputs. I don't think for it, it thinks for itself, based on its code. But I'm guessing that's not what you mean.

Is it because it's got code written by man that it's not "for itself"? But I have genetic code written in me that controls how inputs (sensory organs) are interpretted and what is output. I didn't write that code, it was written through my evolutionary process. So is that really me thinking for myself? Or is that me thinking for my parents who made me, who is thinking for their parents before them? Is it because it arose of organic processes? Well then we're shit out of luck as man-made computers aren't going to form out of organic processes... they're man-made by definition.



I know, I know, I'm running down a tangent of rhetorical nonsense.

But with out a clear definition of what intelligence is, we're shit out of luck.

I always liked the Turing Test for intelligence. I like the simplicity and low bar of it. That low bar shows humility to me.

Why are we so concerned with it being as smart as us? Why can't it be as smart as a dung beetle? (they're pretty smart, they navigate by the milky way)

This post has been edited by lordofduct: 12 February 2013 - 09:18 AM

[jon.kiparsky]

Some useful assumptions which, once accepted, nicely clarify the questions raised above:

Ultimately, all consciousness is composed of subprocesses which are not themselves conscious. There's no principled basis for distinguishing between these processes based on their origins. Turing's test is the most convincing framework yet devised for determining whether an entity "counts as" intelligent. "Intelligent" and "conscious" are very close to being synonymous in this discussion, and they both mean, roughly speaking, the degree to which an entity is an appropriate target for the intentional stance. (ie, intelligent doesn't mean "smart" and conscious doesn't mean "enlightened". Python4President is "intelligent" for purposes of this conversation, though possibly not in other conversations). "Counting as" is the important thing here: there is no sensible way to talk about "really" intelligent/conscious versus "a very good fake".



Note: Nobody is required to accept these, but I find they are quite useful and hard to argue against. I get these from various readings in phil of mind and cognitive science: if you want a good starting place, Dennett's "Consciousness Explained" and Hofstadter's more recent and somewhat fuzzier "I Am A Strange Loop" cover similar ground, and I recommend both for content and also for style. Anyone actually interested in the questions raised in this thread should certainly read those two.

Since we're talking phil of mind, I'll be upfront about my assumptions: I am a thoroughgoing materialist, in that I believe that all "consciousness" is produced by the brain, and there is no metaphysical component to consciousness, or to anything else. While others are of course entitled to their beliefs, I find that strict materialism is the only position which is in any way defensible. I would greatly appreciate knowing where others are starting from, so we don't waste so much time talking past each other.

[gabehabe]

What if, like, we're all computers built by the man, man?

woah

[jon.kiparsky]

lordofduct, on 12 February 2013 - 11:09 AM, said:

Why are we so concerned with it being as smart as us? Why can't it be as smart as a dung beetle? (they're pretty smart, they navigate by the milky way)

What's smart about that? Moths navigate by the moon, and that's why they fly into candles.

[lordofduct]

Moths... smartest creatures on earth.

They know that its all pointless and to just end it now. The purest of nihilists out there. Nietzsche would be proud.

[jon.kiparsky]

Quote

Moths... smartest creatures on earth.

That explains your sig, I guess:

If you want to know how to program, take a moth class, take a lot of moth classes!

[h4nnib4l]

I've always been a little curious as to why (certain groups of) people are obsessed with the idea of androids "waking up". Sweet, I just dropped a fuck-ton of money on a Kara, but the bitch decided she was alive and wanted to experience the world so she peaced out. I hope she's still under warranty. Do you get a refund or a replacement? Doubtful, I'm sure there's a disclaimer for that somewhere in the paperwork that you have to sign to get one.

And why, at a point in the future where we can make Karas in the first place, would they not simply have created an advanced genetic algorithm that, with a simple but involved process of narrowing down the purchaser's concept of beauty, would they not be made-to-order with an appearance that is allowed to develop (somewhat) naturally within the constraints of the purchaser's identified concept(s) of beauty? Seriously, it would be really annoying trying to identify your own Kara at social events. Sure, SHE would be able to identify you from a great distance, by appearance, voice, energy signature, whatever, but I'm stuck figuring out which android walking around the room is MY sex slave... er.. personal assistant.

[lordofduct]

And then they'll have a machine to test if someone is a Kara or not, and distinguish them from the general populous.

And those who illegally jump ship from Mars to Earth will be hunted down and disposed of.

I know what I'm going to do for a job in this future!

A Blade Run.... garbage man.

[WolfCoder]

If you can't objectify women, build women out of objects?