If lusting over well dressed robots is wrong, I don't want to be right!

Check your organic privilege!

>>375277
Rude! Rude to the organically challenged among us.

I'm just gonna leave these here.

Real question: What happens when AI gets to the point where it can accurately stand in for a real human? What happens when we're able to develop synthetic skin for robots to make them more lifelike?

I'm a bit concerned about what will happen to human relations at this point. What does it say for humanity when we stop learning how to deal with real people, and just put robots in their place instead? I am also interested in writing a story about it, so consider this research.

>>375429
http://www.crunchyroll.com/time-of-eve

From wikipedia:
"In the not-too-distant future, androids have come into common usage. Rikuo Sakisaka, who has taken robots for granted for his entire life, one day discovers that Sammy, his home android, has been acting independently and coming and going on her own. He finds a strange phrase recorded in her activity log, "Are you enjoying the Time of Eve?". He, along with his friend Masakazu Masaki, traces Sammy's movements and finds an unusual cafe, "The Time of Eve". Nagi, the barista, informs them that the cafe's main rule is to not discriminate between humans and androids. Within the cafe, androids do not display their status rings, and, when patrons depart, the door is automatically locked for two minutes to prevent someone from following them to discover their true nature."

Not really an answer, but a surprisingly interesting short series. I think that when actual Sentient A.I. becomes a thing, it's mere existence will be a less intrusive and invasive thing than the forces that brought it into being.

>>375434

Hmm, thanks. I'll try to make some time to watch that this week.

I sort of get skeptical of the whole sentient AI thing whenever my computer crashes just trying to load my home page or something of equivalent stupidity.

>>375434

>In the not-too-distant future

Next Sunday, AD

>>375449
Oh, like you've never gone to the kitchen then completely forgotten why you went to the kitchen.

>>375429
I kinda agree with >>375449

To me, it almost seems like technology is going backwards at times, as the more advanced things become, the more potential there is for errors to occur. Having a fight with the blu-ray player proved that not too long ago. I think we're still a long, long, LONG way from robots ever gaining any kind of independent thinking, if such a thing is even possible. I'd say it's more likely we'll make contact with aliens first.

>>375429
The thing that gets me is that if you think about it we still have a slave economy but we replaced the slaves with machines or outsourcing which is the same damn thing.

So to me it seems the first, saddest and most disgusting role human like AI will be applied to will be slavery, in fact a short time ago a robot came out which was advertized as being trainable and cheaper than Chinese labor.

Similarly cybernetics will have numerous stigmas once the initial curiosity wears off, "How do we know that athlete doesn't have a cybernetic leg?" and such travails.

>>375469
>I'd say it's more likely we'll make contact with aliens first.

"In December 1995, the Hubble Space Telescope was pointed at a blank area of the sky in Ursa Major for ten days. It produced one of the most famous astronomy pictures of modern times - the Hubble Deep Field Image. A part of it is shown here. Almost every object in this image is a galaxy typically lying 5 to 10 billion light years away. The galaxies revealed here are all shapes and colours, some are young and blue, whereas others are old, red and dusty. The Hubble Space Telescope has also produced two other similar pictures: the Hubble Deep Field South in 1998 and the Hubble Ultra Deep Field in 2004."

We're a LONG way off considering voyages to Mars, which is essentially on our door step, are voyages measured in YEARS.

AI or far more likely cybernetics is FAR closer and FAR more likely to actually be achieved in our lifetime, point in case we already have cybernetic arms capable of running off the brain's electrical impulses.

>>375473
http://newventurist.com/2012/10/a-glimpse-of-the-future-baxter-the-robot/

https://www.youtube.com/watch?v=Al5RhaJgxxU

>>375449
>>375462
>>375469
Robots are imperfect because we're imperfect, I guess.

I don't know if I want robots to ever become sentient. We have enough problems in our lives without robot uprisings, or zombies becoming a real thing, or running into malevolent aliens.

>>375473
>So to me it seems the first, saddest and most disgusting role human like AI will be applied to will be slavery

Until it develops proven real independent thinking and "emotions", which, again, is still super unlikely, no matter how appealingly sci-fi it may be, can you really call it "slavery"? Is your refrigerator, toaster, washer, or car a slave?

>"How do we know that athlete doesn't have a cybernetic leg?" and such travails.

X-rays.

>We're a LONG way off considering voyages to Mars, which is essentially on our door step, are voyages measured in YEARS.

Who's to say the aliens won't come to us, or haven't already?

>AI or far more likely cybernetics is FAR closer and FAR more likely to actually be achieved in our lifetime, point in case we already have cybernetic arms capable of running off the brain's electrical impulses.

Developing high-tech and realistic AI? Sure. But essentially creating a race of sapient machines that could count as another intelligent species on this planet? When everyday modern technology is still all buggy and error-prone? Not very realistic.

>>375477
>Is your refrigerator, toaster, washer, or car a slave?

You know for a fact that's not even remotely the argument I'm making, good point on the X-ray but you can see what I'm talking about.

>Who's to say the aliens won't come to us, or haven't already?

Because that's as unlikely as us getting to them any time soon.

>Not very realistic.

In some respects yes it's not a realistic goal but on the other hand it's very possible in the same way life likely exists somewhere else in the universe.

It's something very possible and in that respect realistic but fairly far off if at all possible.

>>375478
I think it's cool, yes, but it still seems insanely unrealistic. Unless we start making some sort of organic machines, then you might be talking. But if you're going that route, you might as well just work on cloning with cybernetic implants instead.

>>375480
You're falling into the fallacy of thinking life and sentience are special things that only organic life can recreate. There's no reason a computing system of sufficient complexity shouldn't be do anything the organic brain can. Everything you have ever thought or felt was nothing more than chemicals in your brain being used to trigger the same sorts of on/off switches and relational transactions that a computer does when it calculates, just messier, (at present) higher bandwidth, and with more storage capacity.

You've got a complicated API that has had eons to develop, but much of what's come about in those eons is either unnecessary or actively works against your well-being, just not enough so as to prevent you from breeding.

>>375487
I direct your attention to the phrase "of sufficient complexity".

There may not ever be a need to generate purely artificial life. The Turing Test is not actually a standard by which to truly define artificial life, it is a standard by which the machine is complex enough that it will fool most humans. And even then, how many of us have tried to chat up an IRC bot? The ELIZA program is also worth looking up, for the curious.

With that in mind, there's a real question whether "true sentience" will ever be necessary when "I-can't-believe-it's-not-sentience" is just as good and they're never going to go rogue on you because they're just very complex imitations. Certainly there's very little corporate incentive to program the coffee machines with the possibility of going on strike.

>>375501
Even the argument of "TRUE sentience" versus "I CAN'T BELIEVE IT'S NOT SENTIENCE" is still just semantics, though. You can't even prove that other humans are sentient. We just take it on faith that they are because we know ourselves to be and we extrapolate that other humans must also be since they otherwise appear to be the same as us.

At that point it just comes down to a question of philosophy, with no knowable answer. Once it's able to mimic sentience well enough to call its nature into question, it's not longer fair to question its right to call itself a living, thinking being unless you make humans take the same tests.

>>375487
I'm not ruling anything out, but just because it's possible doesn't mean it's probable.

Here's a thought. While we're debating whether or not artificial creations will ever develop a sentience to match our own, what about animals? They've been around forever, and we have yet to find another species of animal that's on our level, although some come close. Even a "Planet of the Apes" scenario seems a little more likely to happen than Skynet suddenly becoming a reality.

>>375509
It takes animals millions of years to undergo a change like that, whereas computers double their processing power every two years. I don't think any animal that's not already sapient is going to become sapient before artificial life does.

That said, there's a pretty strong case to be made for the great apes being sapient.

I'm the guy who sparked this discussion. Thank you. I'm reading it all as research!

>>375506
The thing is, even with doubling of processor power, code does not spontaneously generate itself. And the usable level for machines is so far below anything we'd actually need for sapience (fyi, not the same thing as sentience, semantics may be bullshit but without that bullshit none of this is possible in multiple senses).

You're under the assumption that a sufficiently complex simulation of a person would in turn become a "real" person, given enough time to stew in its' own head. There is no reason that this would ever be true, and animals are actually a good comparison here. Out of the "stew" of the millions of creatures on this planet, we're the only creatures capable of contemplating a notion like "I think, Therefor I am". This doesn't stop Dogs from dreaming, Whales from singing and Monkeys from practicing prostitution (and gay sex), but higher order function is not needed for higher order activity. And there's some very good reasons to avoid fully autonomous artificial constructs.

Roland here, from Halo 4, for example, is capable of flying starships. Starships equipped with really big guns. It is shown that, as an AI, he has a set of built-in "command phrases" that override his behaviour. Now, he is able to rebel against those command phrases, but this is largely for dramatic effect. In a real world scenario, if the system with hooks in Fire Control, Life Support, and just about every other important system on the ship, if that system tells you to "take a hike", you have just screwed yourself before the ship ever gets off the ground.

All machines are tools of man, but to build true AI would be to flout that notion, to build a "purposeless" tool with no other intent than to mimic our own free agency. That's problematic. The moral quandary of "inventing" a race just to enslave it is a lot more trouble than it's really worth. It's much more likely that AI would resemble Dog from Half-Life 2. Which is still pretty neat, but it's not exactly Legion or EDI.

>>375535
>All machines are tools of man, but to build true AI would be to flout that notion, to build a "purposeless" tool with no other intent than to mimic our own free agency.

Sounds kinda like a modernized version of the original Frankenstein's monster.

>The moral quandary of "inventing" a race just to enslave it is a lot more trouble than it's really worth.

Anyone here who is interested in this idea should check out the 'Second Renaissance' episodes of Animatrix, as that's basically what they're about and they are quite good.

>>375506
>Even the argument of "TRUE sentience" versus "I CAN'T BELIEVE IT'S NOT SENTIENCE" is still just semantics, though. You can't even prove that other humans are sentient. We just take it on faith that they are because we know ourselves to be and we extrapolate that other humans must also be since they otherwise appear to be the same as us.
We already have simulations that are indecipherable from real people: Fictional Characters and False Identities.

sentience=/=sapience
sentience is merely a matter of self-recognition and ability to separate that sense of self from senses and perception
many animals seem to be able to do that, to discern an identity. that's not so difficult.
the hard thing to define is sapience. at what point is an intelligent creature sufficiently intelligent and aware that it has exceeded some threshold and can achieve far beyond its physical form through high level thinking?

we also are too caught up in the ways humans experience the world. imagine what the world must look like for a rat whose primary senses are smell, hearing, and touch? what would that be for a robot? how would that change how we think of self?

now please continue.

>>375553
I don't mean to kill anyone's ideas about what AI may or may not be. I am merely commenting on the now with respect to principles of software design and "how we'd do it". For example, we don't have code that writes other, fully function code yet. We do, however, have code that designs functioning hardware, and development environments that generate repetitive tasks for us. In the future, if we write code that writes code that writes code, all with a specific purpose, who's to say that the end result might not be defined as a new life form?

I tell you what I think will never catch on though? Human-like robots. Like cyborgs, I think that'll work. But the uncanny valley factor on some of those maid robots they show off every year in japan is off the chain. Like, I could deal with Wall-E, I can deal with Asimo.

Hey who wants more semi-related mindless television?
QI XL I13 - Intelligence

>>375564
>For example, we don't have code that writes other, fully function code yet.
Huh? Yes we do. I daresay any programmer who has worked on any relatively large programming project writes self-writing code extensively....or else is extremely inefficient at coding. Writing code that writes your code for you is more or less the gateway between being a beginning programmer and being an intermediate programmer.

Also, that's pretty much the entire point of the LISP programming language.

>>375565
Well, obviously I don't know as much as I thought then. Still though, I feel there is a distinction between the dynamic content generated by LISP program, versus actual programs to write programs, as it were. Although, looking it up on wikipedia, LISP is at the forefront of AI research, which makes sense given its' objective nature.

I think I'll link a few AI/robot related things I find neat:

http://blog.opencog.org/ Seems like an interesting attempt at making general intelligence.
http://www.willowgarage.com/blog | http://www.ros.org/wiki/ ROS is designed to help robot makers/programmer share code instead of starting over every time they get a new robot. Originated at Willow Garage, which makes some nice machines.
http://stevegrand.wordpress.com/ This guy made the artificial life game Creatures and is currently working on another one called Grandroids, though he doesn't post too much about it publicly.
And then I was going to link some other things but I can't find the links for most of them at the moment, so I'll just leave at that and go to bed for now.

>>375565
Hmmm no, very few aspect of professional programming require that sort of thing at all.

>>375798
"Pro" and "Good" are not synonymous. Most professional programmers I've met are pretty shitty and inefficient programmers. Their software probably wouldn't be so buggy and slow if they wrote self-programming code the way the Machine God intended.

>>375802
what sort of argument is that? Nobody fucking care about people you've met, and even less how you judge them.

>>375807
Just like no one cares about >>375798 because being used rarely is entirely unrelated to the question of whether or not such code has been written at all yet, other than in fact confirming that it has been written at all.

>>375808
"people you know", no matter ow many, don't make "facts". especially when it's just 'they do it zrong because they don't do like me" as the sole reason to bring it in the conversation.

>>375809
Anyone who has used products by Adobe, Macromedia or Autodesk, tried to use QuickTime or iTunes on a PC (or a Mac, for that matter) or used Microsoft programs or OSes made before like 2005 knows damned well that professional programmers often write inefficient, bloated, and buggy code.

>>375807
>>375808
>>375809
This is actually a big problem in the disparity of education between various institutions that teach code. In most real-world scenarios, the level of coding ability and the adherence to proper form, as well as "cleanliness" of code, can really vary from programmer to programmer.

There is also, again, the aspect that not all jobs require the level of skill that goes into making a functional A.I. or even a chat bot. If you just start with a dictionary, say English, and attempt to map all the relevant meanings and juxtapositions of all the words, you'd have A) a database significantly larger than most for a single program and B) and unreasonably high amount of combinations and therefor, transactions on that database, to the point where real performance issues may become apparent. A lot of the modern gear we have (say in the last 5 or so years) is powerful enough that a certain amount of "overhead" in memory and cpu processing can be handled without any real performance loss. But even then, and even with the physical limit of current tech, next gen stuff is trying to focus on Quantum Computing and crystal cpus. At least, if the raving lunatics I've been listening to are on the ball.

>>375809
Just so you know, "they do it zrong because they don't do like me" was the entirely unrelated part I was referring to. "Yes we do." was the fact.

I think I'll leave a few more pictures here.

>>375953
dat bimbodroid

>>375953
Oh right, I was going to post more than one image. Also, I was going to see if there was another thread for robots that *aren't* made to look fairly-but-not-quite like sexy humans. For now though, pic related.
Oh wait, I already posted the related image in an image thread (>>369801
), and I'm too tired to choose another one. Let's see if I remember to post more robots tomorrow, then.

Alright, here are cute robo-girls.
Sorry I do not seem to have a supply of hansom robo-guys.

A few of these images are of actual robots, or drawings of them anyway.
For instance, this one holding the underwear here is an HRP-4C, developed by AIST, is shown here dancing along with some humans:
HRP-4C Miim dance demonstration
It's also shown able to walk on some sidewalk.
HRP-4C Walking Outside
And this later video shows her demonstrating a more human-like gait.
http://youtube.com/watch?v=YvbAqw0sk6

>>376120
Well somehow a mangled that last video, which is terrible because it's the one with the most robot butt.
HRP-4C Miim's Human-like Walking

Anyway, another actual robot is the Femisapien, made by Wowee toys, which can dance and make a not-so-great attempt at fencing:
FEMISAPIEN Features
FEMISAPIEN Features
It was also released in Japan as E.M.A., short for "Eternal Maiden Actualization" and specifically marketed as a robotic girlfriend.
http://www.technovelgy.com/ct/Science-Fiction-News.asp?NewsNum=1712
エマ (Eternal Maiden Actualization) PV

>>376120
Or wait,is that a swimsuit? I just presumed otherwise since I figured she wasn't waterproof.

Also, here's another version >>376122 since I have it.

>>376123
For some reason I'm pointing out that "version" makes it sound like another version of the same picture when it is another drawing of the same class of robot, despite what I meant probably being obvious.

While I'm doing that, have a Fireball Drossel gif

FT (Female Type)
Female Type Part 1