U.S. CitiesCity-Data Forum Index
Go Back   City-Data Forum > General Forums > Science and Technology
 [Register]
Please register to participate in our discussions with 2 million other members - it's free and quick! Some forums can only be seen by registered members. After you create your account, you'll be able to customize options and access all our 15,000 new posts/day with fewer ads.
View detailed profile (Advanced) or search
site with Google Custom Search

Search Forums  (Advanced)
 
Old 03-18-2012, 09:48 PM
 
4,533 posts, read 6,924,233 times
Reputation: 2307

Advertisements

Quote:
Originally Posted by Woof View Post
I agree that robotic weaponry could fall into the hands of extremists such as the US Army, and even worse terrorist groups. I wasn't thinking so much about programing errors causing problems as intentional stupidity by humans, but now that I think of it, there's this from about 5 years ago: Robot Cannon Kills 9, Wounds 14 | Danger Room | Wired.com
Thanks Woof. I appreciate your exchange of views which stimulates thought. The link you provided is a classic example of how some things that seem to be great ideas can and sometimes do unexpectedly go terribly wrong. It doesn't have to involve intentional stupidity. It's that people are subject to mistakes and errors, in this case programming which is input by humans.


Quote:
Yes, if it's developed, it will inevitably be used - unless the UN becomes so strong as to put up as many safeguards as possible. It would have to be like nuclear weapons, which are after all more dangerous than AI robots (unless the AI robots could grab control of the entire internet in a bid to control the human race - I think that movie the Matrix had a similar theme? I missed that when it came out, been meaning to watch it.)
Even if the UN mandates specific regulations to safeguard against abuse, there are those in the world who would ignore it. Nuclear weapons are clearly dangerous. But more dangerous than AI robots? It depends on how the bots are used. If AI can be developed to equal or exceed human intelligence, then who knows what can happen if it's directed (programmed) in less than peaceful ways. There are already those who thumb their nose at the UN.

As I said earlier, Issac Asimov drew up the inital idea for Laws of Robotics, with the intention of ensuring that robotics will be used for peaceful purposes. It's an excellent model. The reality is that not everyone will agree or abide by such regulations.

The videos I posted earlier about QBO shows these bots are able to remotely share and learn from each other using Cloud computing. Of course, they'd have access to the Internet.
Cloud computing - Wikipedia, the free encyclopedia

HowStuffWorks "How Cloud Computing Works"

With the scenario of an out-of-control advanced AI network, such a system wouldn't necessarily need to rely on nuclear weapons to be dangerous. It's possible it could disrupt communications, shut down power grids, satellites, medical technology, food supply, etc. I think it would be fair to say that a highly advanced and autonomous AI system would likely manage to set up security systems to prevent it from being shut down.

We are increasingly reliant on network computer systems. I enjoy technology, but as we advance in certain areas, we still need to think things though carefully rather than assume we can take everything for granted with no risk of future consequences. We do have intelligence, albeit not perfect, but we have to use it the best we can.

Admittedly, such a futuristic scenario is sheer speulation. I think it's pretty clear that the military would like to use autonomous robots as long as they can be controlled. As it is, some very deadly war bots have been developed. I'm not certain how many have actually been deployed, although there are a good number of them in the field, primarily for defense and surveillance, but can be used as offensive weapons. For example, SWORDS bots can be equipped with a variety of different weapons, including firearms. They're used primarily for defense, but they could just as easily be used for offense. It just depends on who has the controls. While war bots are indeed remarkable works of technology, most tend to be rather clunky. Now, imagine a flexible autonomous AI android war bot. Why stop there? How about 1000's of them? We don't have them yet, but it could happen. So what do some of today's bots look like?


BigDog Overview (Updated March 2010) - YouTube


SWORDS Robotic Weapons Platform - YouTube


Future Military Robots - YouTube



Quote:
The last sentence I wrote was concerning my mystical beliefs, which are based partly on my experiences and partly on other people's. I wouldn't expect you to share them. I believe the Multiverse (one could call it "God") is sentient. The infinite ocean that is "God" explores his own creation through the minds of biological brains and little drops of himself called "souls". I'd say there's no reason that process couldn't work through artificial intelligences also. Just my private speculation.
Oh, okay, then you weren't talking about cyborgs. So then you think that if an android is intelligent enough, able to think and act autonomously, and able to replicate itself, it would then qualify as a technological life form?
Reply With Quote Quick reply to this message

 
Old 03-20-2012, 12:15 PM
 
Location: Pueblo - Colorado's Second City
11,739 posts, read 18,317,370 times
Reputation: 3858
Quote:
Originally Posted by NightBazaar View Post
I came across this recently, and thought you might appreciate it. Of course, this is only fiction, but it does make one stop and think about the future of robotics, or to be more precise, androids. What if a humanoid machine began to think it was not a machine, but was someone who's alive?

This was made for the Sony PlayStation 3, but as far as I know, it's just a concept demo that could be used as a model for video games, motion pictures, etc. Most attempt at creating human-like robots, or tey to create realistic computer generated characters often fall into the "Uncanny Valley" because while they sort of resemble humans, they tend to have jerky or unconvincing movement and expressions. They don't look alive. Below is a link that shows an example of an Uncanny Valley character.


These vids are best viewed in full screen and high def.

This vid demonstrates the Uncanny Valley.


Heavy Rain: the Casting - YouTube




This is an example that seems to have pretty well removed the Uncanny Valley feeling.


Quantic Dream's Kara - YouTube
This is a interesting video. I think its a little dramatic but overall I think that is the direction AI is going and to be honest I would have no problem dating a robot like that and since I am gay there would be no reason for me not to. I mean its not like I am going to have kids anyway and if I could find a AI that I was attracted to and got a long with why not? Of course I would have him look like a hot male 21 yo surfer or soccer player.

Here is a example of what I would want my AI to look like:



The link: Your source for everything sexy

Last edited by picmod; 01-10-2014 at 01:55 PM..
Reply With Quote Quick reply to this message
 
Old 03-20-2012, 02:26 PM
 
4,533 posts, read 6,924,233 times
Reputation: 2307
Quote:
Originally Posted by Josseppie View Post
This is a interesting video. I think its a little dramatic but overall I think that is the direction AI is going and to be honest I would have no problem dating a robot like that and since I am gay there would be no reason for me not to. I mean its not like I am going to have kids anyway and if I could find a AI that I was attracted to and got a long with why not? Of course I would have him look like a hot male 21 yo surfer or soccer player.
Yes, it's dramatic, although the main idea was that an android (nevermind if it represents a female or a male) with the same programming as all the others somehow deviated from the factory specifications and presumably became one of the the first to think of itself as alive, not just a machine. The flaw apparently being in one of the chips. When it realized it was about to be rejected and disassembled because of the flaw, it became afraid because it wanted to live. In this scenario, we can presume that this particular android had developed its own sense of consciousness and self-identity.

The main reason I posted both vids was to show an example of the "Uncanny Valley" which characters are suppose to resemble humans, but still look stiff and unrealistic - in other words, sort of creepy. They look cartoonish and artificial. With the rendering of Kara appearing much more realistic, the contrast to "Uncanney Valley" renderings is quite impressive.
Uncanny valley - Wikipedia, the free encyclopedia

In the post with the vid of the interactive kid, the kid doesn't look or move in a realistic manner. That's an example of the "Uncanny Valley". What made the difference about the rendering of Kara, was that a real actress was used as the model, and (as I recall), thousands of dots were applied to the actress as points to design a computerized frame with more realistic appearance and motion, such as movement of the eyes, facial expression, smooth motion of walking, mouth when speaking, hand and arm gestures, etc. The large number of dots serve as markers which are used by a computer to scan the actress and recreate the positions on a computerized wire frame. Later on, other details such as hair, eyes, teeth, skin texture and color, were added in over the 'wire frame'.

It's one thing to accomplish such renderings for computer animations. It's not quite the same to accomplish for robotic androids, although there have been some animatronic exceptions although they aren't very complex in a wide range of motions. Actual android bots still look artificial.

Here are some short vids of animatronic babies that seems somewhat realistic. Keep in mind that they don't have the same range of motions and expressions as a much older human. Still pretty impressive though.



Animatronic Premature Baby - YouTube


Animatronic Baby by Millennium FX - YouTube


ANIMATRONIC BABY - YouTube



Here's a bot that looks creepy (Uncanny Valley).


Japanese Crawling Robot - YouTube
Reply With Quote Quick reply to this message
 
Old 03-20-2012, 04:22 PM
 
Location: Pueblo - Colorado's Second City
11,739 posts, read 18,317,370 times
Reputation: 3858
Interesting videos and while I agree we have a ways to go the key remains exponential growth. Because of that we should have AI androids like the ones we talk about by next decade. Then I can have my dream bf!
Reply With Quote Quick reply to this message
 
Old 03-20-2012, 11:39 PM
 
Location: where you sip the tea of the breasts of the spinsters of Utica
8,313 posts, read 10,627,990 times
Reputation: 7988
Quote:
Originally Posted by NightBazaar View Post
Oh, okay, then you weren't talking about cyborgs. So then you think that if an android is intelligent enough, able to think and act autonomously, and able to replicate itself, it would then qualify as a technological life form?
yes

The Japanese have had quite realistic looking robots for a coupla years now, only very slightly jerky in movements.


Female Android Geminoid F - YouTube
Reply With Quote Quick reply to this message
 
Old 03-21-2012, 10:28 AM
 
4,533 posts, read 6,924,233 times
Reputation: 2307
Quote:
Originally Posted by Woof View Post
yes

The Japanese have had quite realistic looking robots for a coupla years now, only very slightly jerky in movements.
I don't doubt improvements continue to be on-going, but as you noted about those jerky movements, you're not going to mistake it with a human. It's those little jerky and odd movements, especially pertaining to facial expressions and movement, that continue to keep these devices in the "Uncanny Valley". Most of these bots tend to be animatronics. I suspect the one in the vid is tethered to power and control cables. It's incapable of functioning independently. Such matters could eventually be resolved, but I think the biggest hurtle that's not likely to be forthcoming any time soon is that of complete independence. What I mean is that androids would most likely be network connected (perhaps like cell phones, gps, wifi, etc.) to remote central memory storage banks, drawing on information as needed.
Reply With Quote Quick reply to this message
 
Old 03-21-2012, 02:07 PM
 
Location: Pueblo - Colorado's Second City
11,739 posts, read 18,317,370 times
Reputation: 3858
Quote:
Originally Posted by NightBazaar View Post
I don't doubt improvements continue to be on-going, but as you noted about those jerky movements, you're not going to mistake it with a human. It's those little jerky and odd movements, especially pertaining to facial expressions and movement, that continue to keep these devices in the "Uncanny Valley". Most of these bots tend to be animatronics. I suspect the one in the vid is tethered to power and control cables. It's incapable of functioning independently. Such matters could eventually be resolved, but I think the biggest hurtle that's not likely to be forthcoming any time soon is that of complete independence. What I mean is that androids would most likely be network connected (perhaps like cell phones, gps, wifi, etc.) to remote central memory storage banks, drawing on information as needed.
You forget about the impact of exponential growth. Because of it we are talking years not decades.
Reply With Quote Quick reply to this message
 
Old 03-21-2012, 02:30 PM
 
4,533 posts, read 6,924,233 times
Reputation: 2307
Quote:
Originally Posted by Josseppie View Post
You forget about the impact of exponential growth. Because of it we are talking years not decades.
So what? Even if it's years and not decades, that has nothing to do with complete independence (along with self-awareness). I agree with you - I don't think it's a matter of decades elther. I think it would be much longer. However, even if computers reach a point of containing more information and are able to process massive amounts faster and more accurately than humans, what exactly is that suppose to mean? That machines will somehow transform into a mechanical form of life? Optimus Prime to the rescue!
Reply With Quote Quick reply to this message
 
Old 03-21-2012, 04:41 PM
 
Location: Pueblo - Colorado's Second City
11,739 posts, read 18,317,370 times
Reputation: 3858
Quote:
Originally Posted by NightBazaar View Post
So what? Even if it's years and not decades, that has nothing to do with complete independence (along with self-awareness). I agree with you - I don't think it's a matter of decades elther. I think it would be much longer. However, even if computers reach a point of containing more information and are able to process massive amounts faster and more accurately than humans, what exactly is that suppose to mean? That machines will somehow transform into a mechanical form of life? Optimus Prime to the rescue!
The amount if information computers can process is directly related to their capability. The reason we have the great examples that have been posted today is directly related to the fact computers have advanced at a exponential rate since the first modern computer was built way back in 1890. Had that not been the case most likely we would be using punch cards today. That won't stop now but will be the reason AI advances at a even faster rate in the near future.

Now the interesting philosophical question is will computers become aware or will they just act like they are?

I want to use a example and for my example I am going to post my dream bf (yes I just want a excuse to use him again but I will make a point as well ).

Moderator cut: image removed

I will name him Ollie.

Now when computers get advanced enough (due to exponential growth it will be sometime in the next 10 to 20 years) so Ollie looks like him and talks to me with emotion so I can't tell the difference between him and a real person. The question of Ollie actually being self aware or is he programed to act like he is self aware wont really matter to me because perception is reality and as far I would be concerned he would be a real bf without all the bf issues in today's world of having one.

Last edited by Marka; 01-10-2014 at 12:17 PM..
Reply With Quote Quick reply to this message
 
Old 03-21-2012, 10:11 PM
 
4,533 posts, read 6,924,233 times
Reputation: 2307
Quote:
Originally Posted by Josseppie View Post
The amount if information computers can process is directly related to their capability. The reason we have the great examples that have been posted today is directly related to the fact computers have advanced at a exponential rate since the first modern computer was built way back in 1890. Had that not been the case most likely we would be using punch cards today. That won't stop now but will be the reason AI advances at a even faster rate in the near future.
Sure, computers process information in relation to their capability. BUT.... they have to have the information to process.


Quote:
Now the interesting philosophical question is will computers become aware or will they just act like they are?
Indeed, that IS the big question.


Quote:
I want to use a example and for my example I am going to post my dream bf (yes I just want a excuse to use him again but I will make a point as well ).


I will name him Ollie.


Now when computers get advanced enough (due to exponential growth it will be sometime in the next 10 to 20 years) so Ollie looks like him and talks to me with emotion so I can't tell the difference between him and a real person. The question of Ollie actually being self aware or is he programed to act like he is self aware wont really matter to me because perception is reality and as far I would be concerned he would be a real bf without all the bf issues in today's world of having one.
If "Ollie" should be marketed in the next few years, you might want to check the price tag first. You also might want to be sure all the bugs are worked out so "Ollie" doesn't have an accidental malfunction and turn out like Ash, or wait for a later version of your hot bot.



Alien [1979] - Ash's malfunctioning - YouTube
Reply With Quote Quick reply to this message
Please register to post and access all features of our very popular forum. It is free and quick. Over $68,000 in prizes has already been given out to active posters on our forum. Additional giveaways are planned.

Detailed information about all U.S. cities, counties, and zip codes on our site: City-data.com.


Reply

Quick Reply
Message:


Over $99,000 in prizes was already given out to active posters on our forum and additional giveaways are planned!

Go Back   City-Data Forum > General Forums > Science and Technology
Similar Threads

All times are GMT -6.

2005-2017, Advameg, Inc.

City-Data.com - Archive 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32 - Top