Can a machine become conscious?

Can a machine become conscious?

Other urls found in this thread:

web.media.mit.edu/~minsky/
twitter.com/NSFWRedditImage

>possible ?
yes.
>will it happen in our lifetime ?
no.

No, they have no souls and never will.

Certainly. WE are machines.
I am conscious. I admit I can't prove the same for any of your guys.

When? Don't know. And since I can't prove other PEOPLE are conscious, I can't prove that a machine is either.
We mostly accept that other humans are conscious, based on their actions. Usually easier than assuming "we" are somehow special. Someday, we'll have to make a decision about machines. That will be more difficult since they're "different". Of course, the same could be said about aliens we might someday meet.

>conscious
Allow me to replace that non-real idea with some things that are real:
>Physiology
>Behavior
>Data Processing and Reporting
And yes, you can make a machine have something comparable to our physiology, behavior, and data processing.

...

>WE are machines.
Only stupid people use the broadness of definitions to attempt to prove a point.

I'm arguing against "vitalism" -- the notion that living things are more than their atoms and are capable of feats which "manufactured" things can never duplicate. I thought that meme died, once and for all, with Wohler's synthesis.

Not playing with words.
Easiest way to show something CAN be done is with an example that it HAS been done.

>conscious
what's that?

Has anyone already mentioned the Chinese Room already?

Not that I've noticed.

If your program responds 1 : 1 with inputs and outputs, then sure, everyone will agree that you're talking to the programmer and not the program. But the size of the book/program would absurdly long absurdly quickly to have any sort of effective output. Which means the program has to do more than just go to a lookup table, it has to compute things, the person in the chinese room has to be able to locate and isolate radicals within individual kanji, determine grammar, and assemble his output sentences himself. Even if he's "simply" been told to do those things, it certainly sounds like he's beginning to understand chinese doesn't it?

We don’t even know what makes us conscious. Why would a really complex computer be conscious just because we built it without even knowing what causes consciousness? How do we know some computers aren’t already conscious for some reason we don’t know yet? Until we know what causes consciousness we can’t be certain of how to create consciousness.

Define: soul

You're both conscious and a machine, so yes.

You need to be at least 18 years old to post.

>m-muh belief in qualia, you're underage because I can't make a coherent argument
That's OK, sweetie, you did your best.

Anyone read Life 3.0 by Tegmark? It's really good, has an excellent up-to-date chapter on this very question.

to become conscious machine by definition need to observe its inner state.
all recurrent networks has conscious

>all recurrent networks has conscious
What about no.

yep, the instruction to the man in the room is pretty much "learn chinese".

A recursive factorial function I wrote in Python does not have consciousness.

Meant for

... Or does it?

I guess can stop trying to make a conscious AI now. A bunch of college students did it in com sci I.

Now , all we need to do to reach the singularity is chain some recursive factorial functions together

>trying to make a conscious AI
That's not a very good goal to begin with since "conscious" is a vague buzzword.

>thinks qualia are non-existent

I think we all know who's the biggest cuck here

>If I call him a cuck it'll justify my unjustifiable position
So do these non-physical "qualia" just coincidentally sync up with physical brains despite not having any sort of interaction with them, or do they have some magical form of interaction that conveniently leaves behind zero measurable impact on the brain?
Hmm... almost seems like "they" don't map to an actual part of reality.

Yes all living things are really chemical machines.

Did you just hurt my feelings?
I'm a real human bean!

Yes. Good thing you're too stupid to ask if a Turning machine can become conscious.

>Hmm... almost seems like "they" don't map to an actual part of reality.

thats literally the definition of qualia you fucking brainlet. qualia are the seemingly different subjective perception of external stimuli, therefore you can't qualiify them because they're not static.

>Nobody disputes that "qualia" literally exist
Wrong, read a book.
>you can't qualiify them because they're not static
lol, better start out with something small that has pictures in it though, maybe some Dr. Seuss.

>ad hominem instead of an argument
like clockwork lol

>wants an argument
OK.
>Nobody disputes that "qualia" literally exist
That's not true. Not sure what more I can say on that point except to point out that's not the case. I'm not sure why you would think it is the case since this is one of the most contentious philosophical topics around right now, people argue about it constantly.
>therefore you can't qualiify them because they're not static
Not being static isn't the same as not being quantifiable (you wrote "qualiify but I'm pretty sure you meant "quantify;" if you didn't mean "quantify" though then I'll just point out being static isn't an argument for something not leaving behind any sort of physical evidence despite interacting with a physical thing).

Can submarines swim?

>Not sure what more I can say on that point except to point out that's not the case
i.e. i have no argument
>you wrote "qualiify but I'm pretty sure you meant "quantify;"
you cant do either because its an abstract concept, but the point is blindly refuting things you dont understand is silly.

The machine is just the machine.
The data inside the machine is what holds consciousness.

My body is a life support structure for my brain, the energy patterns in my brain create thought.
A computer is a physical structure to process data, the data patterns in the computer (???)

That thing on the inside of your shoe.

Yes. And they will be more important than humans will ever be.

Of course. That doesn't mean an electronic computer can be though.

>souls

It's actually energy, and there's a lot of it all over the world. But consciousness? We don't even know how and why it is formed or created.

>my way to analyse, store and communicate information is natural so it means I have a soul and machines don't

You fucking retard.

Instead of being hostile, could you please elaborate?

"We don't even know how and why it is formed or created"

Stop talking in the "we" form when you're clearly uneducated about the topic you're talking about.
Not the person you're replying but I'll make an attempt:

What we define as 'conciousness' is a direct result of extremely complex neural networks.

Ever heard of self-learning/hard AI? It are programs that attempt to imitate how the brain works (neural networks). Ofcourse, due to a combination of technological limitations, lack of knowledge and it being a fairly new concept these artificial neural networks are extremely simple compared to the enormous complexity of the brain. That's why creating hard AI at this moment is still impossible, but it's certainly a hot-topic and you can be sure that there's a ton of research and investment going into the subject.

>What we define as 'conciousness' is a direct result of extremely complex neural networks.
No matter how complex I make a program the complexity won’t make it conscious. I might accidentally make it conscious when I design some part of the massive beast, but I know not what part that would be if it’s even possible in a computer program. Not that poster you were talking to but we need to find out what causes consciousness before we go about creating it.

So what's the thing on the outside called?

>i.e. i have no argument
You're complaining I have no argument that "qualia don't really exist" is an actual position people agree with?
I don't understand what you're looking for here. It's just a fact that's an actual position. It's argued for all the time.
>you cant do either because its an abstract concept
I'd agree "qualia" are an abstract concept, which is another way of saying they don't really exist e.g. numbers are abstract concepts and nobody will ever find a number floating around in physical reality. Abstract concepts are useful pseudo-things we behave around.

What do you mean by conscious? My computer is reacting to my keystrokes, but shows no obvious signs of qualia.

Sure.

>no matter how complex I make a program the complexity won’t make it conscious.

Why make baseless assumptions? Are you religious, or is it something you personally believe? (not mocking, honest question)

The most complex programs aren't even remotely close to replicating the complexity of a newborns brain as of now.
It is indeed impossible to create conciousness with traditional programming, which you might be thinking of.
Neural networks are basically "self learning" programs. They learn the same way like humans do, by trial and error. Self learning is also closely related to perception, so you're getting pretty close to the definition of conciousness no?

Seriously, read up on neural networks if you're interested about the subject.

>conscious
No such thing, brainlet.

>Chinese Room
Your brain uses the exact same process. You just sit behing levels of abstraction, pretending to be retarded.

>qualia
This is your brain on phisolophy. It literally doesn't exist.

>phisolophy
philosophy fuck

>It literally doesn't exist.
IF EVERY HUMAN EXPERIENCES IT IT EXISTS. YOU ARE ABSOLUTELY ABSURD. Whatever your subjective reaction to my post is it’s an instance of qualia.

>The most complex programs aren't even remotely close to replicating the complexity of a newborns brain as of now.
That's a bit misleading because newborn brains wouldn't be any easier than adult brains to replicate. It's the stuff we tend to intuit as "difficult" (e.g. advanced / higher level mathematics) that programs can already do very easily and the stuff we tend to think of as "easy" (e.g. everything newborns do) that programs can't do very easily.
Mostly because thinking stuff is "easy" seems to be an adaptive approach to not even beginning to try to think about the given "easy" thing e.g. most of us don't even begin to think about what all is going into all the little details that allow for our bodies to move around without just immediately losing balance and collapsing onto the floor, and most of us don't even begin to think about how we coordinate little mouth sounds strung together into coded proxies for symbolic meaning without immediately stroking out and just making frustrated moaning noises. I think these are the sorts of things where it's more useful not to think about them in most cases than to try to think about them, hence our default behavior is to pretend they're "easy" even though they're really tremendously complicated.

I simply believe we currently have no scientific proof of what causes consciousness. Complexity could be either a cause or a correlation.

>IF EVERY HUMAN EXPERIENCES IT
That's the thing being disputed, you can't use it as an argument for itself. What we really have that can be corroborated with multiple parties as existing are things like *reports* of "qualia."

>EXPERIENCES
Define "experience".
>subjective
Your brain reacts to stimulus. "Subjectivity" is objective and has a material state and representation, just like everything you "experience".

Yes, in the grand scheme my subjective experience is determined by objective laws, but my for some reason my self is experiencing life as a human, not a dog nor lifeless rock. The only explanation is I have a unique qualia. This is undebatable. It is an evident phenomena we observe, but do not yet understand the mechanisms behind. Once we do understand it we have the potential of instantiating it ourselves.

>Yes, in the grand scheme my subjective experience is determined by objective laws, but
No buts, honestly. You still haven't defined "experience" from an objective point of view. Otherwise, you can just take this to

A direct observation of or participation in events

Damn, so everything experiences something all the time. Sure helps your point about the existence of supr speshul meme abstractions.

Semantics arguments aside, you’re still stupid. You can’t make a conscious AI without knowing how to produce consciousness.

But it doesn't exist. You just impose a layer of abstraction over purely physical interactions in the brain and pretend it's something tangable and different when it's objectively not. An AI running on a neural mesh will have exactly as much "consciousness" as a human.

Not soul

What does conscious mean, do you mean observing? The only thing you know for certain observes is yourself. Everything else is just reacting. Your computer doesn't observe you even when you're typing on the keyboard. It's just reacting. As far as you know everyone could be a solely reacting automaton and you're the only thing that observes. But since solipsism is stupid, we're going to assume there are other humans who can observe. How can we make a machine also observe? Given that we can't even prove other humans observe, we can't prove an AI observes. We can only go on faith which is not useful in science. Therefore, the question "Can a machine become conscious?" holds no merit in science and should be discussed in philosophy instead. A more useful question is can a machine pass the turing test?

>without knowing how to produce consciousness.
to produce consciousness we need to feed back output of the network to the input
brain has such characteristics

That could just be a correlation and not a cause.

Get your smart people talk off the internet, asshole.

Anything you want can be anything, op. A man can be a woman so I don't see why a machine can't be conscious or a potato can't be a ballerina.

Only stupid people try to discredit the person making the argument rather than disproving the argument itself

Nevermind souls they fave no free will

Can a conscious become a machine?

take your ego out your ass you are no different than a dog or a rock

I am clearly different.

Probably.

Current best super computers have the processing power of a cockroach.

Human mind runs a lot of bloat wear.

Yes.

Prove that qualia do not exist.

>behavior

Stop trying to sound smart user, you clearly don't have a hand in AI research

That's a pretty pop-sci thing to say. I don't think you can compare the computations done on a computer to those done in the nervous system in such a way that you could make that determination mean anything substantial.

>behavior is a smart sounding word
lol. What non-smart equivalent should we use for "behavior?" Maybe "stuff doing" would be more your speed?
>I only want to hear arguments from AI researchers.
Here you go:
web.media.mit.edu/~minsky/
All the chapters to The Emotion Machine are linked there under the "Some Publications" section.
Marvin Minsky was one of the founders of MITs AI lab. And he didn't believe "qualia" were literal phenomena that exist either (he was the one who coined the phrase "immanence illusion")

What's the matter, are you afraid of having qualia?

Yes. I think integrated information theory of consciousness is the closest we have to an actual theory imo...

What do you mean by consciousness? We can’t even prove qualia exists despite experiencing it. Engineers can’t do stuff with it until we figured out how it works. We can make an AI that passes the Turing test and looks conscious.

In terms of my emotional response to people who believe "qualia" are literally real and need some updated science to account for them I think it's more a combination of disappointment and frustration rather than fear.
I keep thinking the truth about how:
A) "Qualia" existing as an actual phenomenon wouldn't make any sense (e.g. reports of "it" somehow sync with the brain but leave no evidence of interaction on the brain, and that's really another way of saying there is no interaction because any interaction with a physical object necessarily impacts it in an observable way-- no evidence means no impact means no interaction) and
B) The behavior of reporting "qualia" / acting in reference to "qualia" can be explained without appealing to some magic new science in terms of old fashioned classical physics and (useful) false belief / behavior (and physiology e.g. increased blood pressure)
Is something I could get "qualia" believers to realize, but this belief "qualia" are unassailable, raw, immediate, definitely real things that you can't possibly ever deny seems to keep a lot of people from ever accepting it.

If you deny qualia you deny your own real experience, including your study or use of science. So basically nothing is real and you become a solipsist.

"Qualia" not being literally real doesn't make the physical world not real.
That's like saying because a calculator doesn't have "qualia" its answers aren't reliable.

>We can’t even prove qualia exists despite experiencing it.
We can't prove it exists ''for other people''. However, as a French philosopher observed centuries ago, it is quite trivial to prove it exists ''for you''.

Consciousness can't be tested for, it's a philosophical concept. If you mean a machine that has the same outputs as a human would for every conceivable input, the answer is not yes.

The brain is a very complex mechanism, one we may never understand. But we don't need to understand it to replicate its functionality. Consider neural networks and machine learning, they're black boxes no human can comprehend anymore, despite being created by humans, yet they still work.

That’s not like saying that at all.

Not from a sollipsitic perspective. Is sollipstic experince (not report) invalid?

>as a French philosopher observed centuries ago, it is quite trivial to prove it exists ''for you''.
That "argument" is far from universally accepted. If Descartes had actually "proven" that "qualia" exist for you back then there would be nothing motivating all the modern attempts at thought experiments meant to show "qualia" exist e.g. the "Mary's room" knowledge argument (the idea you could know everything factual about color but still learn something new if you were kept sheltered from color all your life and then shown it) or the "philosophical zombie" argument (the idea you could suppose of people exactly like us in every physical way who were nonetheless different in some extra-physical way by not having "qualia").

Threads like these are why Science and philosophy should stay separate. Just make the damn machine and bitch about its existential nature later.

Why are objective definitions the only validity? Multiple parties are by nature subjective. You cant use that as a validation without theoretical backgroumd for substantiation. Infact multiparty correspondence is only about reliability of observation. On its own thia isnt necessarily the best validity. A single party observation based on better theoretical nackground might be better than a multiparty one and probably is.

You'd need to define machine first.
But just so we're clear, nothing man-made will ever have a consciousness. The best we can do is use artifices to simulate consciousness or intelligence.

>energy
it's actually god and it is everywhere and always

Yes, it is. You're trying to argue because "qualia" aren't literally real that this somehow makes our knowledge of the world less reliable.
Which isn't true because there exist inanimate machines (like calculators) which most people A) wouldn't attribute the capacity of "qualia" to and B) wouldn't consider unreliable just because of A.
As another example, you could have a robot with a mounted camera behaving in response to what the camera takes in, and nobody would claim that robot's camera would be "unreliable" just because the robot wouldn't have an "experience" of visuals.

this
how the fuck is anyone expecting to get an answer to this question if we don't have any idea about human consciousness and OP wants to know about machine consciousness