• Welcome to Smashboards, the world's largest Super Smash Brothers community! Over 250,000 Smash Bros. fans from around the world have come to discuss these great games in over 19 million posts!

    You are currently viewing our boards as a visitor. Click here to sign up right now and start on your path in the Smash community!

Social Link checked the thread. Wow! This is a nice thread!

A_Phoenix_Down

Banned via Warnings
Joined
Jul 16, 2014
Messages
852
The thing with AI is the fact that it has been created without feeling. If you were to create a machine dedicated to killing people do you believe it would experience regret? One of the most powerful feelings a normal human being would experience?

That's what separates our intelligence from a machine. The machine makes decisions based on what was encoded into its "brain" while people make decisions that rapidly change and adapt to the situations that come before them. A machine is built (by humans) with specific purpose. We were built with general purpose.
 
Last edited:

FSK

Smash Ace
Joined
Feb 22, 2009
Messages
547
Location
Norway
I know that there has been some recent-ish interest in analog AI as opposed to digital. Which one are you talking about?
Most likely the AI they teach in a computer science degree. Which is digital since its programmed on a computer.

What do you mean when you say analog AI?

The thing with AI is the fact that it has been created without feeling. If you were to create a machine dedicated to killing people do you believe it would experience regret? One of the most powerful feelings a normal human being would experience?

That's what separates our intelligence from a machine. The machine makes decisions based on what was encoded into its "brain" while people make decisions that rapidly change and adapt to the situations that come before them. A machine is built (by humans) with specific purpose. We were built with general purpose.
A human wouldn't feel regret if they thought they were doing the right thing. Machines can be programmed to adapt to situations that come before them. I don't follow you when you say we were built for a general purpose.

and as a phd in engineering cybernetics I agree with Drigo, AI is dumb and has a long way to go.
 
Last edited:

Zelkam

Resident Wizard
Joined
Oct 8, 2014
Messages
881
Location
Knoxville, TN
Switch FC
1068-8431-0431
Personally, I don't think we'll ever see the day that AI becomes advanced to the point of self consciousness, but if we did I would wager that they would be put on the same moral standing as animals. Right now if a dog were to attack someone there wouldn't be a trial or anything, they would just put the dog down. I have a feeling that's how machines would be treated as well. No matter how advanced AI becomes, it'll always lack the one thing that set humans apart from everything else - a soul.
 

Fox Is Openly Deceptive

Smash Detective
Administrator
BRoomer
Joined
Sep 22, 2008
Messages
3,969
Location
WinMelee, Australia
The thing with AI is the fact that it has been created without feeling. If you were to create a machine dedicated to killing people do you believe it would experience regret? One of the most powerful feelings a normal human being would experience?

That's what separates our intelligence from a machine. The machine makes decisions based on what was encoded into its "brain" while people make decisions that rapidly change and adapt to the situations that come before them. A machine is built (by humans) with specific purpose. We were built with general purpose.
But if the AI is self aware and can learn, then assuming there is nothing controlling it, it can have its own determinative force. It no longer becomes a matter of it being whatever we make it to be, no more than people can make other people what they want them to be anyway (debatable, though I don't want to go into it).

If an AI can think for itself and is self aware, what is stopping it from reflecting? Isn't reflecting just thinking about past experiences? Isn't reflecting a small step away from experiencing something similar to regret?
Most likely the AI they teach in a computer science degree. Which is digital since its programmed on a computer.

What do you mean when you say analog AI?



A human wouldn't feel regret if they thought they were doing the right thing. Machines can be programmed to adapt to situations that come before them. I don't follow you when you say we were built for a general purpose.
Essentially the thing with digital AI is you're restricted to 1's and 0's, but analog AI can store much more than this in multiple states. It's much more similar to the way nature operates, though digital is simpler. Recently there have been some break-throughs with it. Some Aussies claim to have developed the equivalent to an artificial brain cell that works just like ours do.


Edit:
Personally, I don't think we'll ever see the day that AI becomes advanced to the point of self consciousness, but if we did I would wager that they would be put on the same moral standing as animals. Right now if a dog were to attack someone there wouldn't be a trial or anything, they would just put the dog down. I have a feeling that's how machines would be treated as well. No matter how advanced AI becomes, it'll always lack the one thing that set humans apart from everything else - a soul.
Ah, but arguably, if an animal exhibited enough intelligence and self awareness, wouldn't you start to feel uneasy about treating it like an animal? That's why I used the alien example.
 
Last edited:

DarkDeity15

Smash Lord
Joined
Oct 16, 2013
Messages
1,662
Location
Edison, New Jersey
NNID
DarkDeityLink015
Im sorry, @ DarkDeity15 DarkDeity15 . Im 29, so you are really a kid to my eyes... maybe kiddo is just too friendly? If you are not confortable with this, I will not do it again. In other hand, i do have good grammar... maybe I make some mistakes, but i can deffend myself in english environment D:, I'm german, you know...
I don't want, by any means, that you don't try to win... but the odds are that you would be destroyed... just take it easy and learn.


Derzeit leben in Düsseldorf, aber Chile ist wärmer in menschlicher Hinsicht.
Btw, right now my city is a bit cold... but we need some rain, so i think it is ok.
Yeah it's actually my bad. I reacted so badly to that because I thought you meant that in a negative way.

Speaking of AI, I wish CPUs in Smash learned and adapted similar to how high level players would. And no, the Amiibo BS doesn't actually count since they still can't do things that real players can, plus there's all that equipment crap to worry about. It would be a neat training tool for when you can't find someone to play. Especially for folks like myself.
 
Last edited:

Zerker

Smash Apprentice
Joined
Feb 20, 2015
Messages
162
Location
Burien, Washington
NNID
AlexZerker
Speaking of AI, I wish CPUs in Smash learned and adapted similar to how high level players would. And no, the Amiibo BS doesn't actually count since they still can't do things that real players can, plus there's all that equipment crap to worry about. It would be a neat training tool for when you can't find someone to play. Especially for folks like myself.
That would actually be incredibly useful for training offline
Guess we can only dream at the moment, though.
 

BaPTraPLaP

Smash Apprentice
Joined
Jan 4, 2015
Messages
113
Location
State College, PA, U.S.A.
NNID
BaPTraPLaP
Yeah it's actually my bad. I reacted so badly to that because I thought you meant that in a negative way.

Speaking of AI, I wish CPUs in Smash learned and adapted similar to how high level players would. And no, the Amiibo BS doesn't actually count since they still can't do things that real players can, plus there's all that equipment crap to worry about. It would be a neat training tool for when you can't find someone to play. Especially for folks like myself.
The amiibo actually gains bonus stats once level 50 regardless of if you gave it equipment. While the amiibo does tend to pull off combos. In a fighting game, incorperating an A.I. that does combos should be no surprise. They just turn into buffed level 9's with good reactions. But they are still abusable basically go to edgeguard against one and it is dead. They will air dodge without a jump and won't be able to recover. At least with my Mario one.
 
Last edited:

A_Phoenix_Down

Banned via Warnings
Joined
Jul 16, 2014
Messages
852
Most likely the AI they teach in a computer science degree. Which is digital since its programmed on a computer.

What do you mean when you say analog AI?



A human wouldn't feel regret if they thought they were doing the right thing. Machines can be programmed to adapt to situations that come before them. I don't follow you when you say we were built for a general purpose.

and as a phd in engineering cybernetics I agree with Drigo, AI is dumb and has a long way to go.
General means a broad range. We all set our own purposes because we are able to do so. The difference between us feeling regret for tough actions is the simple fact that we are able to determine whether it was right or wrong. A machine doesn't necessarily do that.

@ Fox Is Openly Deceptive Fox Is Openly Deceptive I know where your coming from. That's a pretty big if. If a machine learned how to feel and speak for its OWN actions, then the machine would be as much of a human as we are.

Humans have the ultimate blessing of free will. Machines do not.

If machines developed All of this or if we truly found a way to accomplish this... I would think that would be a scary time for us as human beings.
 

FSK

Smash Ace
Joined
Feb 22, 2009
Messages
547
Location
Norway
@ Fox Is Openly Deceptive Fox Is Openly Deceptive ,

these questions you are asking are on the basis that we would actually develop AI which can mimick or even be considered functional individuals in society. Therefore I'm wondering why would we do that?
 
Last edited:

Elessar

Nouyons TO
Joined
Apr 20, 2010
Messages
2,624
Location
Paraguay
NNID
Veritiel
3DS FC
3711-8466-0515
I agree with Zelkam. While it could be argued that dolly was cloned out of necessity, scientists have an amazing record of doing things just to see whether it can be done or not.
 

Naroghin

Smash Ace
Joined
Nov 5, 2013
Messages
684
Location
Rives Junction, MI
NNID
Naroghin
Humans have the ultimate blessing of free will. Machines do not.

If machines developed All of this or if we truly found a way to accomplish this... I would think that would be a scary time for us as human beings.
Computers do already have this to a limited degree and it's been developing over the last few years. Just last year there was a study where robot helpers with a more "free" AI (that is, took risks without knowing the outcome, tried new things, improvised, made "human" mistakes) were given to people to complete a task, while helpers with more "traditional," robot-y AI's (precise, accurate, did not act unless outcomes were "certain," and would stop to observe and analyze before taking any actions) were given to others to help with the same things. The study found almost unanimously the people preferred the more human AI's, but that's not the point; the point is that we can already program some aspects of freewill into computers. Just observe this robot 6 years ago improvising with a human performer, not only following tempo, key, and style cues, but also trying out ideas and making "mistakes."

Of course, I can choose whether or not to improvise, while the robot can only choose what to improvise within certain parameters, but isn't that only one order away? When I am acting out "freewill," is it not within a set of parameters based on current, arguably limited human understanding of the definable and affect-able universe? The question isn't whether or not we'll be able to grant human level freewill to computers, but whether or not we've found a further level of freewill by that time and what implications that could have regarding pretty much everything.

#CTHULUSDREAMSARECONTROLLINGUSALL
 

A_Phoenix_Down

Banned via Warnings
Joined
Jul 16, 2014
Messages
852
Computers do already have this to a limited degree and it's been developing over the last few years. Just last year there was a study where robot helpers with a more "free" AI (that is, took risks without knowing the outcome, tried new things, improvised, made "human" mistakes) were given to people to complete a task, while helpers with more "traditional," robot-y AI's (precise, accurate, did not act unless outcomes were "certain," and would stop to observe and analyze before taking any actions) were given to others to help with the same things. The study found almost unanimously the people preferred the more human AI's, but that's not the point; the point is that we can already program some aspects of freewill into computers. Just observe this robot 6 years ago improvising with a human performer, not only following tempo, key, and style cues, but also trying out ideas and making "mistakes."

Of course, I can choose whether or not to improvise, while the robot can only choose what to improvise within certain parameters, but isn't that only one order away? When I am acting out "freewill," is it not within a set of parameters based on current, arguably limited human understanding of the definable and affect-able universe? The question isn't whether or not we'll be able to grant human level freewill to computers, but whether or not we've found a further level of freewill by that time and what implications that could have regarding pretty much everything.

#CTHULUSDREAMSARECONTROLLINGUSALL
Exactly as you have stated, it's only improvising based on what he is doing, merely "simulating" a free will. Extremely impressive nonetheless. However, as soon as that machine decides to stop playing with that human performer all together is when we really hit that border.

It's crazy to think how close to that we really are to that.
 
Last edited:

Naroghin

Smash Ace
Joined
Nov 5, 2013
Messages
684
Location
Rives Junction, MI
NNID
Naroghin
Exactly as you have stated, it's only improvising based on what he is doing, merely "simulating" a free will. Extremely impressive nonetheless. However, as soon as that machine decides to stop playing with that human performer all together is when we really hit that border.

It's crazy to think how close to that we really are to that.
It sometimes feels like I'm merely simulating a free will, as well. The problem with freewill is that we assume that since we have a concept for it, we must have it.

I personally don't fear the development of AI and robotics, as I believe there is a much deeper and more convincing difference between humanity and AI that we won't discover until we take the concept of AI to its logical conclusion. I think once we hit that wall we may be closer to understanding ourselves.

BTW it's conversations like this that continually remind me why I love the Link social over all others. You guys are pretty awesome, freewill or not.
 

A_Phoenix_Down

Banned via Warnings
Joined
Jul 16, 2014
Messages
852
It sometimes feels like I'm merely simulating a free will, as well. The problem with freewill is that we assume that since we have a concept for it, we must have it.

I personally don't fear the development of AI and robotics, as I believe there is a much deeper and more convincing difference between humanity and AI that we won't discover until we take the concept of AI to its logical conclusion. I think once we hit that wall we may be closer to understanding ourselves.

BTW it's conversations like this that continually remind me why I love the Link social over all others. You guys are pretty awesome, freewill or not.
My problem with all of it is if AI develops free will, they will become too human. If there's one thing history has shown, it's that humans seek conquest and control.
 

8MAN

Smash Apprentice
Joined
Apr 19, 2015
Messages
166
Location
Seattle, WA
NNID
FourOhFour
3DS FC
4871-4261-8409
My problem with all of it is if AI develops free will, they will become too human. If there's one thing history has shown, it's that humans seek conquest and control.
Agreed, once development of AI reaches the "free will" stage, we'll probably all live in a dystopia, where humans are all eliminated or enslaved by the machines.
 

Vonzar the Soulrender

4th Dimensional Horror
Joined
Oct 8, 2008
Messages
819
It sometimes feels like I'm merely simulating a free will, as well. The problem with freewill is that we assume that since we have a concept for it, we must have it.

I personally don't fear the development of AI and robotics, as I believe there is a much deeper and more convincing difference between humanity and AI that we won't discover until we take the concept of AI to its logical conclusion. I think once we hit that wall we may be closer to understanding ourselves.

BTW it's conversations like this that continually remind me why I love the Link social over all others. You guys are pretty awesome, freewill or not.

I agree here. Humans breed animals for specific personality traits and temperments all the time. Why do we think that our thoughts, personalities and affinities are truly our own? Who I am right now could easily entirely be the product of my genetics responding to environment.
 

Fox Is Openly Deceptive

Smash Detective
Administrator
BRoomer
Joined
Sep 22, 2008
Messages
3,969
Location
WinMelee, Australia
I don't think it's a matter of 'if' so much as it's a matter of 'when' we'll see artificial intelligence indistinguishable from us. Most predictions see this happening within our life times, but then predictions have been wrong before (where is my hover-board?). There are many people that see no reason why analog AI couldn't become conscious, while digital AI, i.e. what everyone is used to, seems to have no hope of that. The current trend is to simply copy what nature does, and so far on a small scale it has been working. People are seeing the possibilities of combining synthetic biotechnology with analog AI to create new conscious life forms. We'll just have to wrap our heads around that, because it's coming. I think that the biggest obstacle stopping it from happening at the moment is the laws that we've set in place, because it's wierd and we don't want to have to deal with it, not to mention that it's arguably dangerous, perhaps extremely dangerous.
 

Rizen

Smash Legend
Joined
May 7, 2009
Messages
14,887
Location
Colorado
Everything computers or AI do is always based on formula. There is no 'true randomness' that can can be generated by computers. For example on this site it says:
https://www.random.org/
"Perhaps you have wondered how predictable machines like computers can generate randomness. In reality, most random numbers used in computer programs are pseudo-random, which means they are generated in a predictable fashion using a mathematical formula. This is fine for many purposes, but it may not be random in the way you expect if you're used to dice rolls and lottery drawings.

RANDOM.ORG offers true random numbers to anyone on the Internet. The randomness comes from atmospheric noise, which for many purposes is better than the pseudo-random number algorithms typically used in computer programs."
This is why AI is different than true free will.
 

Fox Is Openly Deceptive

Smash Detective
Administrator
BRoomer
Joined
Sep 22, 2008
Messages
3,969
Location
WinMelee, Australia
Everything computers or AI do is always based on formula. There is no 'true randomness' that can can be generated by computers. For example on this site it says:
https://www.random.org/
"Perhaps you have wondered how predictable machines like computers can generate randomness. In reality, most random numbers used in computer programs are pseudo-random, which means they are generated in a predictable fashion using a mathematical formula. This is fine for many purposes, but it may not be random in the way you expect if you're used to dice rolls and lottery drawings.

RANDOM.ORG offers true random numbers to anyone on the Internet. The randomness comes from atmospheric noise, which for many purposes is better than the pseudo-random number algorithms typically used in computer programs."
This is why AI is different than true free will.
That would be digital though. Digital is much more focused on doing only what is programmed.
 

Lawz.

The Original Deadpool Fan
Joined
Dec 29, 2008
Messages
1,056
Location
Pennsylvania
NNID
Lawz_Fearless
I don't think it's a matter of 'if' so much as it's a matter of 'when' we'll see artificial intelligence indistinguishable from us. Most predictions see this happening within our life times, but then predictions have been wrong before (where is my hover-board?). There are many people that see no reason why analog AI couldn't become conscious, while digital AI, i.e. what everyone is used to, seems to have no hope of that. The current trend is to simply copy what nature does, and so far on a small scale it has been working. People are seeing the possibilities of combining synthetic biotechnology with analog AI to create new conscious life forms. We'll just have to wrap our heads around that, because it's coming. I think that the biggest obstacle stopping it from happening at the moment is the laws that we've set in place, because it's wierd and we don't want to have to deal with it, not to mention that it's arguably dangerous, perhaps extremely dangerous.
I see that Back to the Future reference Foxxy.
 

link7

Smash Lord
Joined
Jan 14, 2013
Messages
1,081
Location
Steilacoom, Washington
NNID
Kidlink77
3DS FC
2234-7140-8163
So after some consideration, I'll be heading to a small local tourney today. Thanks for pushing me @Volcanic Doomfire
Tourney is allowing Pokémon Stadium 2 and Weegee's Mansion as legal stages. I'm insta banning Weegee, but I can't wait to bombslide across the icy terrain in PS2. It's stupid fun, and Link goes flying.

@Zane the pure I'm going to add you on my Wii U. Would you mind uploading some vids on Youtube I'll send you? They're of me playing WA's current best Link, Tallguy. Intense stuff.
 

Rizen

Smash Legend
Joined
May 7, 2009
Messages
14,887
Location
Colorado
I got some good wifi games in this morning. Which is nice because I was doing terrible last week or whenever.
 

Himura Kenshin

Smash Journeyman
Joined
Feb 2, 2015
Messages
248
It's not guaranteed. The only thing that is close to being close to being close to being close to guaranteed is jab 2 to spin attack.
 
Top Bottom