At what point could an AI become human?

[Tm_Carbon]

I was wondering on what you guys thought about AI becoming human. If the AI reached a level where it could have feelings would it then be human? If you classify humans by their soul then what makes up a soul and would it be possible for an AI to get one? Just curious on what people think about an AI becoming human and whether they think it's possible or not. 

[Polyvalent]

neural networks already exist, but considering the fact that we still dont understand 100% of the brain and how it functions, I'd say it would be incredibly difficult for a computer to perfectly emulate a human.

[Tm_Carbon]

1 minute ago, Vizzion said:

neural networks already exist, but considering the fact that we still dont understand 100% of the brain and how it functions, I'd say it would be incredibly difficult for a computer to perfectly emulate a human.

But hypothetically if we could make a computer that would perfectly emulate a human would it then be human?

[Polyvalent]

Just now, Tm_Carbon said:

But hypothetically if we could make a computer that would perfectly emulate a human would it then be human?

 

no, it would be a computer that can perfectly emulate a human.  when you emulate GBA games perfectly on your PC, does that make your pc a GBA by extension?

[Tm_Carbon]

2 minutes ago, Vizzion said:

 

no, it would be a computer that can perfectly emulate a human.  when you emulate GBA games perfectly on your PC, does that make your pc a GBA by extension?

Even if the computer had free thought and had feelings of its own? If I can perfectly emulate GBA games on my pc then yes I would by extension say my pc is a GBA.

[Polyvalent]

2 minutes ago, Tm_Carbon said:

Even if the computer had free thought and had feelings of its own? If I can perfectly emulate GBA games on my pc then yes I would by extension say my pc is a GBA.

 

I have to disagree.  You can make an AI with free thought, feelings, and emotions, but at the end of the day its still an AI, not a human.  And your PC does not become a GBA by extension, your PC just has the same capabilities but with far different properties.

[Satisfoxy]

I for one accept our new robot overloads.

[Tm_Carbon]

2 minutes ago, Vizzion said:

 

I have to disagree.  You can make an AI with free thought, feelings, and emotions, but at the end of the day its still an AI, not a human.  And your PC does not become a GBA by extension, your PC just has the same capabilities but with far different properties.

Alright I can see how you got that. I do have a different question for you however. What would you say makes a human, human? 

[Dash Lambda]

Not in the way you're saying.

I absolutely believe that computers can develop to the point where they can demonstrate human-like sapience, and I would say that at that point the computer is human, but not a human.

 

Humans are a biological species, and computers are obviously not part of that species. A computer can "be human" in the same sense that a human can "be athletic."

EDIT: Oh, and also, the poll is pointless if the only option is "yes."

[Tm_Carbon]

1 minute ago, Dash Lambda said:

Not in the way you're saying.

I absolutely believe that computers can develop to the point where they can demonstrate human-like sapience, and I would say that at that point the computer is human, but not a human.

 

Humans are a biological species, and computers are obviously not part of that species. A computer can "be human" in the same sense that a human can "be athletic."

Alright I have to admit I did not think about the human species when I wrote this. If a computer was able to have the qualities of a human without being part of the species would you acknowledge it as human to the point that it has rights and should be treated like the species of humans?

[Dash Lambda]

13 minutes ago, Tm_Carbon said:

Alright I have to admit I did not think about the human species when I wrote this. If a computer was able to have the qualities of a human without being part of the species would you acknowledge it as human to the point that it has rights and should be treated like the species of humans?

Like I said, a computer can (or, at least, I believe one can given more development) effectively "be human." To exclude a truly intelligent machine from the decency we show normal humans would be closed-minded and a tad barbaric.

[AntonChigurh]

A computer cannot "be a human". A computer can become truly intelligent and can truly become human when it

1.) Learns how to have feelings

2.) Learns how to ask questions

 

Otherwise, it is nothing more than a very complicated neural network. Also, to answer the original question, yes a computer can become sentient. Not right now, but at some point in the future technology (hopefully) will have advanced so far that we will be able to create true artificial intelligence. But that's assuming that we don't reach a point where technology can't advance any further and Moore's law becomes a reality first.

[MadyTehWolfie]

If it skwaks like a bird but is a cat is it a new type of bird?

[Sauron]

Define what being human means... it's not like that has been one of the greatest unanswered questions in the history of mankind...

 

Either way current advancements in AI work in a completely different way from the human way, it's reallty not comparable - apples to oranges.

[BuckGup]

I can see in a couple years we achieve smarter AI that learns more complex functions on it's own. But for AI to be self aware we are going to have other hurdles outside of technology. People will be moral crazy calling it a human, or people scared of humans become obsolete. 

[Sauron]

43 minutes ago, N1ghtshade said:

A computer cannot "be a human". A computer can become truly intelligent and can truly become human when it

1.) Learns how to have feelings

2.) Learns how to ask questions

 

Otherwise, it is nothing more than a very complicated neural network. Also, to answer the original question, yes a computer can become sentient. Not right now, but at some point in the future technology (hopefully) will have advanced so far that we will be able to create true artificial intelligence. But that's assuming that we don't reach a point where technology can't advance any further and Moore's law becomes a reality first.

A (non AI) computer can be easily programmed to have "feelings" in response to certain situations and ask questions. You'll have to be a bit more specific than that if you want to define when a computer will be considered sentient.

[AntonChigurh]

47 minutes ago, Sauron said:

A (non AI) computer can be easily programmed to have "feelings" in response to certain situations and ask questions. You'll have to be a bit more specific than that if you want to define when a computer will be considered sentient.

What I mean is to ask questions of its own accord. To truly be inquisitive. Out of all the intelligent creatures on earth, humans are the only ones truly able to ask questions. Even what we consider to be highly intelligent animals such as apes lack that ability. They have no idea there is anything outside their current realm of knowledge, and so they cannot ask questions. If an AI were truly human, it would know that it does not know everything, and that it would need to ask questions to gain more information.

 

edit: I think this video breaks down AI very well

 

[dalekphalm]

4 hours ago, Tm_Carbon said:

I was wondering on what you guys thought about AI becoming human. If the AI reached a level where it could have feelings would it then be human? If you classify humans by their soul then what makes up a soul and would it be possible for an AI to get one? Just curious on what people think about an AI becoming human and whether they think it's possible or not. 

AI cannot become human, because a human is a very specific,  biological organism. 

 

An AI could become a sentient, thinking, self-aware being with human-like intelligence and emotional range though. But that still doesn't make it Human.

 

That would be like finding a Cardassian in space. They have human-like intelligence. They have emotional range, feeling, thought, and they are sentient. But they aren't human, even if they share all the same traits as us.

[straight_stewie]

4 hours ago, Tm_Carbon said:

I was wondering on what you guys thought about AI becoming human. If the AI reached a level where it could have feelings would it then be human? If you classify humans by their soul then what makes up a soul and would it be possible for an AI to get one? Just curious on what people think about an AI becoming human and whether they think it's possible or not. 

It is probably possible that a computer could one day become sentient, however, as I will cover below, there are many very difficult hurdles to overcome. Most of the hurdles involve the evolution of humans, language, and thought. Just to be technical though, a human follows the following taxonomy: Animalia->Chordata->Vertebrata->Mammalia->primates->hominidae->homo->homo sapiens.

A human is an evolved (or created, or both) animal. A computer can never be that.

 

3 hours ago, N1ghtshade said:

A computer cannot "be a human". A computer can become truly intelligent and can truly become human when it

1.) Learns how to have feelings

2.) Learns how to ask questions

Those two criteria have nothing to do with sentience at all. Having feelings is not a basic requirement of sentience by any logic, and asking questions is only a byproduct of being aware of ones self and the things around it. Believe it or not, the most common belief is that we cannot truly tell whether a being is sentient or not, which leads me to the most important point of a topic like this:





We, as humans, lack the language and ideas necessary to describe intelligence and sentience. At our present stage of evolution we cannot know if something is intelligent or not. We cannot describe sentience, we cannot describe a general intelligence. We have no idea  how this works or why this works. Therefore, we cannot purposefully build a machine that has sentience or general intelligence. 

[themctipers]

10 hours ago, Satisfoxy said:

I for one accept our new robot overloads.

Me too ( ͡° ͜ʖ ͡°) 

[SansVarnic]

By creating AI to the point of emulating a human it would essentially be an android.

 

Data, the one and only ...

 

[yathis]

You talking right now, 5 yrs ahead, 10 years ahead

 

How far ahead are you talking about?

[SansVarnic]

IMO technology wise, it would be at the very least 20 years before we see a level of awareness in AI's to emulate human passably.

[Polyvalent]

11 hours ago, Tm_Carbon said:

Alright I have to admit I did not think about the human species when I wrote this. If a computer was able to have the qualities of a human without being part of the species would you acknowledge it as human to the point that it has rights and should be treated like the species of humans?

 

ultimately if what you say happens and computers are human in every aspect except their biology (synths basically) I'd say they can have rights and whatever but for the safety of all humans, it would imo be best to "shackle" any type of AI so that they cannot go full skynet.  perhaps a failsafe hard-coded into the AI

[SansVarnic]