Will advances like AI and nanotechnology make humans obsolete?

[Ultima]

Think about this one. AI and nanotechnology are both "God" technologies, in that the "complete" versions of either will enable that country to essentially play "God". However, AI will easily be smarter than us and could render us obsolete. A supersmart AI with control over nanotechnology WOULD be a god in essence.

Imagine this scenario. The first AI is created with the intelligence of the average human being. This AI is capable of expanding its mind and becoming ever smarter. Let's say that it takes the AI 4 years to double its intelligence. It's now as smart as two people. Now that it has double the smarts, it only takes it 2 years to again double its intelligence to that of 4 people. After 7 years it has the power of 8 people, at 7.5 years 16 people, at 7.75 years 32 people, and this continues exponentially as the AI gets more and more powerful. By 8 years the AI's intelligence is nearly infinite. Where would an almighty AI with such capabilities leave us humans? Would there be a point in doing anything at all? A supersmart AI would know everything, could do anything. With nanotechnology, it could turn China into a giant smiley face and turn the moon into a square if it so desired. What would be the future of the human race under those circumstances? Would that be the end of our voyage?

 

[syzygy]

what kind of intelligence do you speak of ? computational ? emotional ?

computers can't feel, can't read a person's mood, can't judge the rational behind religion, the need for creativity, don't
know squat about inspiration, and a gazillion other nuances that cannot be programmed but need to be intuited by the
natural powers of persons. etc.

it seems to me that computer (ai) can compete with the raw ability of a human brain but cannot be taught anything
about mind, the metaphysical, and the like.

 

[glenn1]

Will advances like AI and nanotechnology make humans obsolete?

No, because you can't program uniquely human qualities like creativity and ambition into a machine.

 

[0roo0roo]

well, are you sure about that? whats magical about ambition? but seriously, we're no where near anything like decent ai. the amount of knowledge/enviromental awareness/logic just walking around in a city is mind boggling.

i'm still waiting for a stupid pc to respond to voice commands. not that via voice bullsh*t, but startrek level

 

[glugglug]

Not ALL humans. But a large percentage of people on these forums are useless and obsolete already!

 

[Ultima]

Originally posted by: syzygy
what kind of intelligence do you speak of ? computational ? emotional ?

computers can't feel, can't read a person's mood, can't judge the rational behind religion, the need for creativity, don't
know squat about inspiration, and a gazillion other nuances that cannot be programmed but need to be intuited by the
natural powers of persons. etc.

it seems to me that computer (ai) can compete with the raw ability of a human brain but cannot be taught anything
about mind, the metaphysical, and the like.

WHY couldn't the AI be taught about these things. If the AI has virtually infinite computational power, it could surely be able to compute those things no? Even if not mathematically, it could simulate a human brain at virtually no cost and then it would be able to understand all those concepts. All our brain is a collection of neurons which also compute, just like AI would do. I don't think there's anything mystical about us that cannot be reproduced by a powerful AI. I think you're missing the whole point of AI when you speak of programming. An AI wouldn't be programmed except for the core seed to get it started. It would learn on its own.

 

[charrison]

Originally posted by: Ultima

Originally posted by: syzygy
what kind of intelligence do you speak of ? computational ? emotional ?

computers can't feel, can't read a person's mood, can't judge the rational behind religion, the need for creativity, don't
know squat about inspiration, and a gazillion other nuances that cannot be programmed but need to be intuited by the
natural powers of persons. etc.

it seems to me that computer (ai) can compete with the raw ability of a human brain but cannot be taught anything
about mind, the metaphysical, and the like.

WHY couldn't the AI be taught about these things. If the AI has virtually infinite computational power, it could surely be able to compute those things no? Even if not mathematically, it could simulate a human brain at virtually no cost and then it would be able to understand all those concepts. All our brain is a collection of neurons which also compute, just like AI would do. I don't think there's anything mystical about us that cannot be reproduced by a powerful AI.

Before we can simulate how the brain works, we first have to understand how it works.

 

[Ultima]

Originally posted by: charrison

Originally posted by: Ultima

Originally posted by: syzygy
what kind of intelligence do you speak of ? computational ? emotional ?

computers can't feel, can't read a person's mood, can't judge the rational behind religion, the need for creativity, don't
know squat about inspiration, and a gazillion other nuances that cannot be programmed but need to be intuited by the
natural powers of persons. etc.

it seems to me that computer (ai) can compete with the raw ability of a human brain but cannot be taught anything
about mind, the metaphysical, and the like.

WHY couldn't the AI be taught about these things. If the AI has virtually infinite computational power, it could surely be able to compute those things no? Even if not mathematically, it could simulate a human brain at virtually no cost and then it would be able to understand all those concepts. All our brain is a collection of neurons which also compute, just like AI would do. I don't think there's anything mystical about us that cannot be reproduced by a powerful AI.

Before we can simulate how the brain works, we first have to understand how it works.

Who said anything about simulating the brain? That wouldn't be a good goal for designing an AI. The last thing we need is an AI with a core designed on greed, ambition, hatred and all those other things wired into us.

I meant that the AI itself could simulate the brain to understand those "human qualities" it it wanted to, and with virtually infinite computational power it wouldn't take very long for it to run the few quadrillion calculations to figure it out

 

[Moonbeam]

Humans are a scream. Most don't seem to notice they are machines. There is always the possibility of a better machine. For billions of years life has been refining an organic program to succeed at the sim called the universe. Only now can we begin to play with the game via genetic engineering. All we need to do is to create a machine that can run a program that can midify the rules and test for adaptive advantage within a sim that modles the universe and then the universe itself.

 

[LunarRay]

Originally posted by: syzygy
what kind of intelligence do you speak of ? computational ? emotional ?

computers can't feel, can't read a person's mood, can't judge the rational behind religion, the need for creativity, don't
know squat about inspiration, and a gazillion other nuances that cannot be programmed but need to be intuited by the
natural powers of persons. etc.

it seems to me that computer (ai) can compete with the raw ability of a human brain but cannot be taught anything
about mind, the metaphysical, and the like.

What is it about the human brain and its 'programming' - the mind, that permits humans to 'feel' that can't be incorporated into the CPU/Memory of this box on my desk here.. Seems to me that there is not an organ or substance in the human brain that can't be harvested and 'Frankensteined' into a computer.

 

[singh]

I think that if we are able to one day "create" true AI, it will be largely accidental.

 

[sandorski]

Not to worry, Bush would confound the AI and Arnold would Terminate the puny machines.

 

[syzygy]

Originally posted by: Ultima

Originally posted by: syzygy
what kind of intelligence do you speak of ? computational ? emotional ?

computers can't feel, can't read a person's mood, can't judge the rational behind religion, the need for creativity, don't
know squat about inspiration, and a gazillion other nuances that cannot be programmed but need to be intuited by the
natural powers of persons. etc.

it seems to me that computer (ai) can compete with the raw ability of a human brain but cannot be taught anything
about mind, the metaphysical, and the like.

WHY couldn't the AI be taught about these things. If the AI has virtually infinite computational power, it could surely be able to compute those things no? Even if not mathematically, it could simulate a human brain at virtually no cost and then it would be able to understand all those concepts. All our brain is a collection of neurons which also compute, just like AI would do. I don't think there's anything mystical about us that cannot be reproduced by a powerful AI. I think you're missing the whole point of AI when you speak of programming. An AI wouldn't be programmed except for the core seed to get it started. It would learn on its own.

a computer language works with 1 and 0. regardless of how many combinations you can make of those
numbers, they are intrinsically limited to mathematical applications, which negates this 'infinite computational
power'. a person's limits are not bound to a 1 and 0. math simply does not exist in human behavior.

a person is susceptible to stimuli long before they understand what development is. infants, toddlers, children
do not understand the effects and ramifications of their experiences unless they learn to reflect, a skill set which
not everyone develops. not being able to understand will result in persons who are violent or some in way criminally
prone with a range of emotional and behavioral issues which they will carry throughout life in varying degress of intensity.

you would have to build your AI with similiar flaws and vulnerabilities that would be essential to its successful functioning.
if you choose to eliminate these complications from the code, then you would be aiming for perfection, which in essence
is an impossibility and no longer defines a human being. there goes your objective.

 

[Compton]

I don't think we know enough about the human brain to answer this question.

I think technology could be used to enhance human life, but the bible beating politicians will undoubtably make that illegal.

 

[CrazyHelloDeli]

If anyone is interested in these concepts, I highly recommend a book by Michio Kaku called Visions

 

[rbV5]

Think about this one. AI and nanotechnology are both "God" technologies, in that the "complete" versions of either will enable that country to essentially play "God"

Supposedly God made the Heavens and Earth in a matter of days...don't think that stuff gets you quite there.

 

[dmcowen674]

Originally posted by: syzygy

Originally posted by: Ultima

Originally posted by: syzygy
what kind of intelligence do you speak of ? computational ? emotional ?

computers can't feel, can't read a person's mood, can't judge the rational behind religion, the need for creativity, don't
know squat about inspiration, and a gazillion other nuances that cannot be programmed but need to be intuited by the
natural powers of persons. etc.

it seems to me that computer (ai) can compete with the raw ability of a human brain but cannot be taught anything
about mind, the metaphysical, and the like.

WHY couldn't the AI be taught about these things. If the AI has virtually infinite computational power, it could surely be able to compute those things no? Even if not mathematically, it could simulate a human brain at virtually no cost and then it would be able to understand all those concepts. All our brain is a collection of neurons which also compute, just like AI would do. I don't think there's anything mystical about us that cannot be reproduced by a powerful AI. I think you're missing the whole point of AI when you speak of programming. An AI wouldn't be programmed except for the core seed to get it started. It would learn on its own.

a computer language works with 1 and 0. regardless of how many combinations you can make of those
numbers, they are intrinsically limited to mathematical applications, which negates this 'infinite computational
power'. a person's limits are not bound to a 1 and 0. math simply does not exist in human behavior.

a person is susceptible to stimuli long before they understand what development is. infants, toddlers, children
do not understand the effects and ramifications of their experiences unless they learn to reflect, a skill set which
not everyone develops. not being able to understand will result in persons who are violent or some in way criminally
prone with a range of emotional and behavioral issues which they will carry throughout life in varying degress of intensity.

you would have to build your AI with similiar flaws and vulnerabilities that would be essential to its successful functioning.
if you choose to eliminate these complications from the code, then you would be aiming for perfection, which in essence
is an impossibility and no longer defines a human being. there goes your objective.

"math simply does not exist in human behavior."

Huh? It's called Irrational Computations

You find a lot of that on here for example.

 

[syzygy]

Originally posted by: dmcowen674

"math simply does not exist in human behavior."

Huh? It's called Irrational Computations

You find a lot of that on here for example.

well, probabilities and esoteric number games are amusements for pointy-headed people.
they don't translate to real-life applications where being wrong can be a source for inspiration,
for example.

 

[Hayabusa Rider]

If the human brain were simple enough to understand, we would be too simple to understand it.

 

[Ultima]

Math actually does exist in human behavior. Get down to it and everything is just mathematical interactions between molecules. A neuron either fires or it doesn't. Looks like 1 and 0 to me. However,The point of AI would NOT be to emulate a human being nor to model it on survival-of-the-fittest evolution. Make an AI like that and it will destroy us. I don't see any reason why you couldn't eventually do it though with advances in technology.

Originally posted by: syzygy

Originally posted by: Ultima


a computer language works with 1 and 0. regardless of how many combinations you can make of those
numbers, they are intrinsically limited to mathematical applications, which negates this 'infinite computational
power'. a person's limits are not bound to a 1 and 0. math simply does not exist in human behavior.

a person is susceptible to stimuli long before they understand what development is. infants, toddlers, children
do not understand the effects and ramifications of their experiences unless they learn to reflect, a skill set which
not everyone develops. not being able to understand will result in persons who are violent or some in way criminally
prone with a range of emotional and behavioral issues which they will carry throughout life in varying degress of intensity.

you would have to build your AI with similiar flaws and vulnerabilities that would be essential to its successful functioning.
if you choose to eliminate these complications from the code, then you would be aiming for perfection, which in essence
is an impossibility and no longer defines a human being. there goes your objective.

 

[Ultima]

Originally posted by: rbV5

Think about this one. AI and nanotechnology are both "God" technologies, in that the "complete" versions of either will enable that country to essentially play "God"

Supposedly God made the Heavens and Earth in a matter of days...don't think that stuff gets you quite there.

Ok, it wouldn't be able to do that, but for all intents and purposes.. those nanites can do practically anything Lots of scientists are scared of "gray goo" scenario in which nanites consume the entire planet...

 

[Ultima]

One thing that crossed my mind about advanced AI though is that wouldn't it eventually ask itself "What's the point?" and come to the realization that it's more efficient to simply not exist and destroy itself?

 

[djNickb]

I had a class on data modeling and knowledge management in my last semester of college. The course was actually quite interesting as we did some modeling using neural networks. Whats cool about them is that they are learning systems, just as a human learns and makes decisions about situations based on previous experiences the neural network does it in the same way. You can use them to find relationships between certain causes and effects that you might not otherwise even think to exist. They're also great for designing decision support systems. Its very interesting stuff to say the least.

 

[djNickb]

Originally posted by: Ultima
One thing that crossed my mind about advanced AI though is that wouldn't it eventually ask itself "What's the point?" and come to the realization that it's more efficient to simply not exist and destroy itself?

"I cannot self terminate"

 

[Moonbeam]

Originally posted by: syzygy

Originally posted by: Ultima

Originally posted by: syzygy
what kind of intelligence do you speak of ? computational ? emotional ?

computers can't feel, can't read a person's mood, can't judge the rational behind religion, the need for creativity, don't
know squat about inspiration, and a gazillion other nuances that cannot be programmed but need to be intuited by the
natural powers of persons. etc.

it seems to me that computer (ai) can compete with the raw ability of a human brain but cannot be taught anything
about mind, the metaphysical, and the like.

WHY couldn't the AI be taught about these things. If the AI has virtually infinite computational power, it could surely be able to compute those things no? Even if not mathematically, it could simulate a human brain at virtually no cost and then it would be able to understand all those concepts. All our brain is a collection of neurons which also compute, just like AI would do. I don't think there's anything mystical about us that cannot be reproduced by a powerful AI. I think you're missing the whole point of AI when you speak of programming. An AI wouldn't be programmed except for the core seed to get it started. It would learn on its own.

a computer language works with 1 and 0. regardless of how many combinations you can make of those
numbers, they are intrinsically limited to mathematical applications, which negates this 'infinite computational
power'. a person's limits are not bound to a 1 and 0. math simply does not exist in human behavior.

a person is susceptible to stimuli long before they understand what development is. infants, toddlers, children
do not understand the effects and ramifications of their experiences unless they learn to reflect, a skill set which
not everyone develops. not being able to understand will result in persons who are violent or some in way criminally
prone with a range of emotional and behavioral issues which they will carry throughout life in varying degress of intensity.

you would have to build your AI with similiar flaws and vulnerabilities that would be essential to its successful functioning.
if you choose to eliminate these complications from the code, then you would be aiming for perfection, which in essence
is an impossibility and no longer defines a human being. there goes your objective.

Ah me, somebody who has one hand on the vine tells us of the limits of human perfectability.

"You limit the scope, you limit the findings" LunarRay