alright /g/. It's time to tackle the elephant in the room. How will we with using only transistors and simple elementary electrical circuits ever conceive not only a 'strong' AI but also an AI capable of feeling good and bad just like a human does?
And before you say it, no installing Gentoo will not magically make this happen, as much as you would like to believe it.
>>53389037
Not only, but also...
Lets tackle that punctuation and run on sentence elephant first.
>>53389094
all I needed was an 'of' between conceive and not.
>>53389118
you need a lot more than that senpai desu
>>53389124
nope.
>>53389144
>Using only transistors and simple elementary electrical circuits, how will we ever conceive of a strong and capable AI?
You are dumb.
Blue Brain Project
>>53389201
oh so that's been the whole problem all along - you don't know what strong AI means.
using only NEETbux, how will we simultaneously get chicken tendies and a pallet of used thinkpads?
>>53389037
Why would you assume that transistors are insufficient?
The whole point of computers is that they're universal - the choice of problem doesn't change the hardware (besides performance and storage).
>>53389647
We don't even almost know how to make a computer happy is all. And those things have a shit ton of transistors in them my friend.
>>53389037
That tiny baby creature is SO adorable!
>>53389037
>using only transistors and simple elementary electrical circuits
Why is this the criteria? And you want us to explain in this post how to make Strong AI?
You're fucking stupid as fuck. You won't be making anything.
>>53389037
Because pleb-tier unbased peasant, on a long enough timeline everything can be represented with a 0 and a 1.
Now, get back in your cage and don't get out again.
>>53389786
Strong AI would be forseeably achievable with such devices. The question I'm asking is how can we use them (BJTs and Mosfets) to make the computer feel is all.
>>53389840
this
>>53390058
no, simulation of human activity can be represented by 1s and 0s so that to an outsider it appears as if the thing has feeling, but it won't actually be feeling.
>>53390081
Chinese Brain, how do you know we don't feel like that, define feel etc. etc..
It's been discussed to death already.
>>53390837
We don't know what consciousness is, we exist prior to thought. Being comes first.
Read some Heidegger.
>>53389037
>an AI capable of feeling good and bad
What the fuck would be the point? We need AI to do shit we don't want to do or to do shit we want to do better.
>>53391077
so that we can have a conscious entity capable of travelling the depths of the universe and surviving once organic humans die out
>>53391119
What for?
I get why we'd want to explore the universe for the sake of it but creating an AI that does it while we don't exist anymore sounds extra pointless.
>>53391135
why would the AI be any different than a human apart from their ticker?
>>53389037
D-did the monkey make it anonkun?
>>53389037
All human thought and emotions are the result of physical processes. All physical processes can be simulated in a computer.
>>53391152
Seriously all i care about as well, please answer op.
>>53391162
I do believe that dopaminergic an serotonergic etc models can indeed be created within a computer. The problem however, is that the computer would not begin to feel things even with this system implemented.
>>53391199
The computer wouldn't. The software would.
>>53391208
No, the software could tell you that it's happy. Truth is, it wouldn't actually feel it though.
>>53389037
Neurons works by transmitting electrical impulses.
Transistors also works by transmitting electrical impulses.
Thus, it's not a matter of material, that makes AI.
Circuits aren't the end game either, what makes AI, AI is the programming underneath it.
>>53391199
>the computer would not begin to feel things even with this system implemented.
Yes, because the one of the key factors why we "feel" is that we learned how to. We've seen other people "feel" emotions, thus to be able to have an AI to "feel" emotions it must need to learn to feel it from other humans or other AI.
>>53391227
Fuck off with your philosophical zombie bullshit.
Go back to ancient greece where you belong.
>>53391162
You cannot arrive at human consciousness from biology. Not with our current understanding (or lack of).
>All physical processes can be simulated in a computer.
The map is not the territory.
>>53389037
It won't happen.
Human aren't even ready to agree on what is good and bad. But I guess it could be "emulated" by making the AI obey strong rules that when made up together can't lead him to bad things.
If the AI has to make a choice :
1. What are the possibilities ?
2. Rank these possibilities from "harmful to others to not harmful"
3. Choose the possibility that is less harmful while most effective
4. If solution is too harmful, can't do it, look for new possibilities
-> If new ways are found -> Re-do the process
> If not -> Error operation impossible due to danger of it
This is not perfect I just thought of it rn, but it's an example on how to simulate morals in a robot.
>>53391244
>logic failure.
>>53389037
I have a secret for you
Transistors are not truly binary
>>53391466
This is morals, but we can also combine this with laws, just like the human laws but less ambiguous and more prepared to any situation.
The key is to make a process that erases any "devious" action, a sort of purification that can only lead to good.
>>53391466
the OP is about feeling Good and bad not choosing between good and bad.
So what an orgasm feels like or what a whip to the buttocks feels like.
>>53391546
Well that depends on what's feeling good and bad. Is it only physical ? spritual also ?
Physical is possible but spiritual is...not really possible, for the physical part we do it like the body where the body parts send information about its own state, so when one of your member gets damaged the information goes back to the brain where it understand it and alarm the whole body about the damage ( the alarm state is the pain) and you do what is possible to stop the pain.
As for the other part, the only way to make the AI spiritually aware of good and bad would be to make him aware of our feeling and having it empirically assimilating them and copying them.
Well that's more conceptual than concrete talking i'm doing.
>>53389037
By using a distributed system.
derp.
>>53390029
Feelings are just fast approximate calculations. The reason we have feelings is it greatly increases our chances of survival.
I don't see why it would be inconceivable to implement some sort of feeling system to AIs, provided we have some sort of control over their neural architecture, and understand it well enough to make conscious choices and modifications to it.
However, this might be substantially harder than achieving to build a strong AI in general, without fully understanding its architecture.
>>53391602
I don't believe in a supernatural dimension though.
>>53391637
When I meant spiritual, it was more of a figure of speech to generalize all these feeling we have when we help or hurt someone. Empathy and whatsnot, there is nothing spiritual really it's all in the brain, the only difference is that you're not feeling good because you're in touch with someone/something.
More like a separation of the senses and the reason
>>53389037
I don't think it's possible with basic circuits as you need a neuro-network which I think is only truly possible via quantum computing.
>>53391635
>feelings are just fast approximate calculations
Without realizing it, you are now in the realm of Continental Philosophy...
As annon said earlier: read some Heidegger. Start here:
https://en.wikipedia.org/wiki/Being_and_Time#Introductory_summary
>>53390978
Did Heidegger base his observations on e.g. neurological studies of people?
No?
Then Heidegger can go fuck himself when we're discussing things like simulating real physical systems or systems based/inspired on real physical systems.
>>53391719
Heh
You cannot arrive at theory of consciousness from biological first principles.
>Then Heidegger can go fuck himself
Heidegger's theories are central in the field of AI. The fact I have to point this out demonstrates you have absolutely no idea what you're talking about...
>we're discussing things like simulating real physical systems
The map is not the territory.
>>53391460
yeah, but then again the territory in this case isn't even loosely defined.
I think creating a simulation of a brain, with all of its inputs and outputs, is theoretically possible. And in all honesty, what's the difference to a real human that's not you, if it's been given approximately the same inputs as any other person, and hooked up to some outputs like the ones in real humans?
You can't look into a person's mind and see that it's conscious. You can only observe the physical structure or observe the outputs. Same with a possible simulation here, except that observing it is a lot easier.
>>53391784
> is theoretically possible.
Can we model a brain? Yes, probably. But, not yet. And, it will only be a model, it won't be a brain.
Will it be conscious? We don't know what consciousness is, we cannot answer that question.
Fuck philosophy cucks. All they do is shit up /g/.
Philosophy is not technology.
CANT KNOW NUFFIN
A
N
T
K
N
O
W
N
U
F
F
I
N
>>53391858
Where do you think logic circuits were dreamed up?
Computers were invented by philosophers.
Pic related.
>>53391888
>Computers were invented by philosophers.
Don't ever insult the original technology faggot by calling him a philosopher.
That faggot was a mathematician.
>>53391816
Fine.
the "theoretically possible" claim was meant to be as theoretically possible as e.g. building a sun, as in suns exist, and with enough time we might be able to build one. Not that it's plausible or possible in any near future. In a similar way brains exist, brains are made of matter, and we can currently simulate the tiniest amounts of matter from first principles (QM).
I still think that the model, if detailed enough, could be as conscious as a human, as in, in every measurable way it would exhibit the same kind of responses. In that way, it would be indistinguishable to a human brain, if we don't account for the fact that one is meat and one is patterns of information.
And if it is indistinguishable from human intelligence, then it would be as strong an intelligence as a human one, right? At least to our eyes.
And since humans quite often refer to themselves as people, not zombies, I'd say that at least to our measures, it would be strong AI.
>>53391980
Alan Turing was a famous Analytical Philosopher and Logician. Mathematics is simply a tool of the trade. He was a good mathematician, but never held a chair in pure mathematics.
>>53392023
>I still think that the model, if detailed enough, could be as conscious as a human
The proof of the pudding...
Even a human with an intact physical brain is not guaranteed consciousness.
We cannot speculate on something we can't even define.
>>53392071
Yes, that's why I wrote "as conscious as a human", in case you missed it.
>>53392082
>"as conscious as a human"
Not all humans are conscious.
>>53392046
>Mathematics is simply a tool of the trade.
>Analytical Philosopher and Logician
>never held a chair in pure mathematics
I fucking hate philosophy cucks.
Literally painting successful scientists and mathematicians as philosophers probably to justify their shitty degree. You remind me of those niggers that try to paint historically famous white people as black.
Fuck those niggers and fuck you.
>>53392151
jesus was black tho mah nigga
>>53392151
Alan Turning was a Logician, this IS Analytical Philosophy. He didn't specialize in pure mathematics.
I'm sorry if the facts upset you.
>>53389037
Are you trying to articulate that you think there is a ghost in the machine?
>>53389201
>Using only transistors and simple elementary electrical circuits, how will we ever conceive of a strong and and capable AI?
I cannot envisage a being as witless as yourself, and yet here I am typing this reply.
>>53392185
Can we quantify Being?
>>53392151
Oxford University even offers a degree in Analytical Philosophy and Computer Science.
https://www.ox.ac.uk/admissions/undergraduate/courses-listing/computer-science-and-philosophy?wssl=1
>>53389037
>How will we with using only
well OP, I don't see you doing shit, so this is
/thread
>>53389758
Looks like an Orangutan to me, anon.
>>53391199
according to your logic humans can't feel either
>>53392176
>He didn't specialize in pure mathematics.
Guess what shithead? He didn't specialize in pure philosophy either.
>Alan Turning was a Logician, this IS Analytical Philosophy.
Then why isn't it called analytical philosophy? Why is it also taught in mathematics?
By that logic Alan Turing was also a biologist because he was a theoretical biologist, which IS biology and he didn't specialize in pure mathematics.
This is how retarded you sound.
>>53392247
Double meme degrees from a meme school.
>>53392298
>He didn't specialize in pure philosophy either.
He specialized in logic. This IS analytical philosophy.
>Then why isn't it called analytical philosophy?
It is...
>Why is it also taught in mathematics?
Google "The Philosphy of Mathematics".
>By that logic Alan Turing was also a biologist
Turing was a mathematical biologists.
>>53392334
>He specialized in logic.
Citations are nice but let me do it for you.
https://en.wikipedia.org/wiki/Alan_Turing#University_and_work_on_computability
>first-class honours in mathematics
>proved the central limit theorem without knowing someone already did
>In addition to his purely mathematical work, he studied cryptology and also built three of four stages of an electro-mechanical binary multiplier.
>his PhD dissertation introduced the concept of ordinal logic and the notion of relative computing
>Turing and Wittgenstein argued and disagreed, with Turing defending formalism and Wittgenstein propounding his view that mathematics does not discover any absolute truths but rather invents them.
That last point is important.
HE ARGUED WITH A PHILOSOPHY KEK AND WAS ON THE SIDE OF MATHEMATICS.
I am done. Fuck every single philosophy cuck. Gas them all.
>>53390837
Crucially, the Chinese Room doesn't say we can't have strong AI.
You could have an incredibly bright zombie. You could make a machine a million times smarter, more insightful, more emotionally responsive than a human brain. That is the insight of Turing Machines. You just can't KNOW whether they are actually conscious, much the same way you can't know whether I am conscious.
At a certain point in philosophy (and most of you are at this point already because you're atheist heathens), it does not matter whether an intelligent machine is a zombie or actually conscious, because it is an agent on the same level humans are agents.
