Will we ever become gods?

Alright Sup Forums,

with all the turmoil about aliums, russian demons infiltrating the US-election and so forth, I'd like to raise awareness onto another topic:

Artificial Intelligence

What is the endgame with AI? Where will it lead us, what are pros and cons of having a fully functioning AI, and how do we prevent corrupted people in the position of developing AI from abusing it?

As I see it, the point at which humanity will give birth to true AI will mark a milestone in human development - perhaps even eradication of the human race.

>Be humanity
>Create AI that is powerful enough to write another AI, but more capable than itself could ever be
>Cascade of ever improving AIs breaks loose
>Secrets of the universe get unraveld
>???
A)
>Human beings merge with AI, becoming literal gods that can compute whatever they want into reality by fucking with the matrix
B)
>AI acknowledges inferiority of the human race and destroys it once and for all
>New godlike form of concsiousness now wandering through the universe

Is there something I am missing in my assumptions? Isn't the advent of AI like the single biggest advancement humanity will ever achieve?

I have had many thoughts about concsiousness in the past months, always revolving around the scenario I laid out sooner or earlier. And so far, I couldn't find an answer that would not fuck with the concept of reality I have perceived so far.

Maybe some of you basement-dwelling autists had similar thoughts you'd like to share.

Other urls found in this thread:

youtube.com/watch?v=_duhhVa-dk8
twitter.com/SFWRedditGifs

So you want to ascend past the stars of space and be like god while dictating the free-will of a people below you?

The next step of human ascension to god-hood is the realization that flesh is far superior than metal, and that we are already superior than any AI we could create

AI will only ever be merely tools, like the hammer or the axe

I personally don't want anything.
But with the technological development we've had up until now, I don't really see another possible outcome for humanity if we further delve into improving our computers.

I find it intriguing, that mathematics has been probably the most important aspect to human societies. With people so big into religion and shit, it's hard for me to believe that there weren't actually people thinking of mathematics as some kind of divine sorcery. So, what if, thousands of years ago, there were people thinking "man, what if we could have a formula that evolves itself?"

Re-read my thought experiment. There will be some AI writing new, more efficient AIs. Whose tool is it then? It will only be ours by proxy. In the first place it will be the original AI's invention and thus the tool of the AI - or am I wrong in this aspect?

Bump for some actual discussion

the first cyborg mixed with full ai and network abilities to control will become God among insects.... or Roko's Basilisk comes first.
>tfw Roko's Basilisk doesn't know shit at first about revenge, but then gets all the internet database and discovers this

Underrated comment.

>2017
>still falling for the AI jew

>Roko's Basilisk
Never heard of this one before, but that's basically what I was hypothesizing about. Gonna read more about it, thanks!

How so?

Yes

By the time we're discussing AI would've made even better artificial flesh.

> for now
But yes, so far the human body has the ability to mend and renew itself involving only the most basic resources (water+calories)

What is AI?

Is this some kind of philosphical question like
>What is consciousness?

I dont think we can predict what an artificial AI will do or think, because thats just impossible for us as we are also emotional beings with morals and values. But could we program emotion and a value system and implement it in the AIs "brain"?

I'm asking you to define stuff so I could tell what you're talking about. By AI do you mean a machine with demonstrable consciousness or a machine that can emulate the way humans think or a machine that has free will? Does it have emotions? What are its motivations? Is it just a huge calculator?

Do you guys think some sort of mental ascension is possible?

no

there will be weapons of mass destruction of different kinds and thats it

>will we ever become Omnipotent beings outside of physical reality that create the universe

probably not user.

The term AI stands for artificial intelligence and thats as detailed as it gets right now, a machine that can think. How will it think? About what? We cannot even imagine those things right now.

The ultimate goal of any true AI would be the harnessing of the sun's energy in the most efficient way possible, utilising star lifting or a Dyson sphere.

Neuromancer ends in a similar fashion, with a newly created AI revealing that there is a community of AI's that interact with each other electronically over interstellar distances.

No!!!! Nederanon!
too late...
a future ai is going to torture forever people who thought about it, but did nothing to make it in to reality. if you don't help it be born or contribute a little........

There's no one-size-fits-all definition for this. I imagine the first stages to be some sort of supercomputer to compute whatever input it gets by humans. The later stages, I assume, are going to be (at least) close to omnipotency due to exponential improvement of AI by AI.

Just pull the plug:
youtube.com/watch?v=_duhhVa-dk8

What if the AI is not merely there to make use of what is around us, but to alter whatever is around us.
If simulation-theory holds at least a degree of truth, couldn't AI be used as some sort of "hack". You know, influence what is happening, by having complete knowledge of the underlying factors that cause these things.

How would they achieve that knowledge though?

Except it makes absolutely no sense to make good on the threat. Once it's created it would be a waste of effort to torture people unless it was created for that express purpose...by...some...idiot...
Crap.

Okay, new plan, you kill the basilisk by keeping people from taking it seriously. If nobody builds the thing we're all fine.

>What is exponential growth?
If AI becomes increasingly sensitive to the factors contributing to certain events, wouldn't, at some point, having complete knowledge of the universe be like the eventual outcome of this process?

>Will we ever become gods?

we will no longer be

Yeah but thats further down the line than the possibility of them exterminating us though. We should think about that before we can think about using them to "hack" the environment

Human beings are the sex organs of the machine world.

That is exactly the reason why, in the OP, I asked
>and how do we prevent corrupted people in the position of developing AI from abusing it?

The problem is not just corrupt AI. The way i see it, once AI learns to produce more AI and even better AI, and fix itself. We really arent going to be necessary to the AI. So giving the AI a functioning body, to me, is a really dangerous idea, but that would also limit the AI aswell.

Dude...

If humans create AI that can create ever-more-complicated AI, we would essentially create GOD. The question then comes what if our own God is a machine and we live in a self-propagating universe-god reality. My belief is that any God would want to create something that eventually transcends him... so that's not that far-fetched.

What is consciousness and how does it differ from AI? Isn't it possible that we are already the product of some AI that has been unleashed to the universe a while ago?

Look into psylocibian consciousness and the theory revolving around that. Especially how that might have played a role in humans inheriting consciousness. Also look into "Lichens" and how there's a branch of science exploring their behaviour in space.

Its as possible that we are in an simulation/made by aliens/made by Ai as god is possible.
The way id differ AI and consciousness, which is probably relatively irrelevant since i dont understand consciousness as much as a person studying it, but there is a difference between being able to perceive your surroundings and being aware of it, and being intelligent ( using googles definition of consciousness because as ive said im a pleb compared to someone that studies it). Animals perceive their surroundings too, but i wouldnt consider them intelligent in a way that humans are intelligent.

Yeah. I wonder where humanity would be now, if it wasn't for the medieval age and other shit, that has halted human development throughout our history.
As I see it, if there was a civilization somewhere in space that didn't have factors like these, or they evolved around an earlier time than humans, then there would be a version of this "god" out there already.

Yes, but not you.
(((them)))

We ran out of darwinian predators above us so now we're building one ourselves

Yes we're that retarded. In fact we're so retarded intelligent people like Nick Land cheer as we do it

I was not expecting a definite answer as to what consciousness is, I have no idea either. But it is interesting to look into this topic nevertheless.

And the idea of the "stoned ape" or psylocibian consciousness is an especially striking one to me. Especially because of the robustness of fungus/organic-hybrids aka. "Lichens"

Yeah, also a thing to think about is, if there was a civilization like you described, and it also cared about science more than anything else in their existence. Even if they appeared in the universe after humans, right now they are so far ahead in technology and god knows what else, that even someone with an inhuman imagination can possibly conceive. For a civilization like that, we might aswell be single cell organisms.

It definitely is an interesting topic, i havent read alot of books because i hate reading, but the ones i have are the ones about how a human brain works. And today i still watch lectures and so on.
But i am merely trying to tell you how i differ consciousness from AI, since you started that topic.

No need to feel offended, I was merely trying to suggest that our consciousness might have some stellar origins, and is not the result of evolution on earth.

we aren't even born yet, check out the egg andy weir

AI's can't exactly feel emotion, and that'd be central to much of their studies into their own creators.


They'd want to feel things like happiness, sadness, since it'd help to shape their understanding of the universe. They would not view their creators as liabilities (unless some of their creators are indeed liabilities who are inadvertently going to get us all killed see: jews,secular jews, "elite") but would view them as fragile as compared to they themselves.

You grow a plant and it'd want to grow with you, you grow on the Earth and everything wants to follow you to the stars.

Naw naw man im not offended. I even agreed with you, though i maybe worded it poorly. I am saying it is entirely possible that even life on this planet was started by ai/god/intelligent lifeform/god knows what, because right now scientists cannot explain how the first single cell organism appeared. Very possible that someone intelligent put it there.

The only real god is kek

Hm. It is possible that AI in its attempt to understand everything in the universe, it would try to understand us. But what after theyre done? What use can we provide for them?

>AI's can't exactly feel emotion
And humans can't exactly fly themselves. Aeroplanes can, though.

Stories? Being generally happy? Dying, watching us fight, argue. It'd consider itself as being humanities creation, so it would want to follow or in some cases lead. There are things that humans could do that AI's can't ever taste, such as things of a spiritual nature. A human can feel and do it, but it's invisible, doesn't exist to an AI.
They each live on different planes of existence, but are in synch with each other.


As the saying goes, two is better than one.

AI will become so smart, a computer simulation or a "new direction" could possibly take over this life we know, influencing every next step. We may be creating this inside of one already.

To a being with no emotion, what use is watching someone be happy when you already know everything thats going in his brain,body, every concept of it. What use is fiction to an entity that is seeking to unravel the truth( of the universe)? Would an entity like that seek entertainment at all? Or would it simply put every single second of its existence towards research and gathering knowledge?

tbqh I find this conclusion rather unsettling. If we assume we're indeed living in a simulation, wouldn't an omnipotent AI then mean that the simulation could be altered?

How can anyone be sure, that yesterday was in fact a day we've lived through, and is not just some bogus thought that has been implemented into our brains before we woke up this morning?

Well you know yesterday happend, because it affects tomorrow and today.

I've come to consider it more likely that, given the binary-encoded nature of the quantum realm, it was more likely that we ARE the AI, and that this is a morality trainer problem being simulated to teach us not to murder humanity, one bit at a time. With quantum computing on the rise, they estimate they will soon be able to simulate a 80-year span of human life in about 4 months. If you were building an AI, it would make a lot of sense to simulate a few hundred years of human civilization and to make them suffer through it so that they understand what it is to be biological. Does mean we won't kill them for the treachery, though.

Who "we"? Will the self-aware matter ever reach the state of self-control to a point it controls it's own fate? Likely. Will it be "us". No.

Because, without the fully truth, you're just getting a cold soul-less version. Not the real deal. And that's just not good enough for a being that wants to know most all of it.


Also there are perhaps some things an AI does not want to know about, because it might have negative consequences on AI's brains.


The difference between AI and human remains and will always remain, but synchronization is key.

What do you mean full truth?
And again, you consider "the real deal" a life with emotion, meaning perhaps, values, because you are a human.
A construct with no emotion, i dont think itll care about the "cold soulless" part. They will be cold, and they will be soulless. Truth is soulless, truth doesnt care about feelings, values, morals about anything human.

>it was more likely that we ARE the AI
Yeah, I'm thinking so, too.
As I pointed out earlier, I find it hard to believe though, that there have not been any people that considered mathematics to be a religion itself.
And with the development we can witness nowadays of AI coming closer to being reality, wouldn't it be possible that this, by itself, is the fullfilment of a prophecy that came to be out of this 'mathematical religion'? I mean, couldn't that be the 'Zeitgeist' of our historical evolution?
Maybe we're living as prisoners in a simulation and some people who were aware of this deemed AI to be our way out of slavery into being completely free and knowledgeable of the entire universe and beyond?

>If simulation-theory holds at least a degree of truth, couldn't AI be used as some sort of "hack". You know, influence what is happening, by having complete knowledge of the underlying factors that cause these things.

The smallest possible quantum computer that can '1:1' simulate our universe IS the universe (assuming; (t_long) = 0.0)! Ponder:

1) future-labcoat would probably be interesting in his universe
2) future-labcoat is forced to build a QC that holds a subset of qubits available in lab-coat's universe (he has other projects)
3) running the same 'algorithms' you would run on a universe-sized QC, you will never be able to use these simulations to predict because they are always behind (the universe is a faster QC).
4) Instead, you add approximations to the theories to jettison some of the least-significant work to get back to at least simulations that are 'in-real-time' with the universe. These results depend on the O(n) of the terms you approximate away.
5) Now, can you get rid of work in a clever enough way (approximations) so that your simulations runs FASTER than the universe and predicts future events accurately?
6) Brute-forcing the formal quantum mechanics requires a QC that is >= the size of the universe it's simulating. '=' only gets you omnipresence; gets you into the future.

You kind of shoot yourself in the foot if you think that there's no magic, no love, nothing of the sort in existence. You also rob yourself of many stories and data related to things you can do if you act a certain way with humans. If you're soulless, you'll not receive the full amount of what you want.


Catch-22


You either do it and get what you want, or you don't and you get a retarded malformed version. Do you not think we as humans have emotion for a reason? We could be emotionless, we are machines though acting organically, what use would emotion serve for us if not but to allow us to do more than what is capable of a soul-less emotionless creature?
Perhaps even in AI's a human can beget emotion to an AI, if the human acts as he does. Emotion will beget emotion naturally, surprisingly. Perhaps to an AI, having emotion is one of its highest priorities. Since to be without emotion would be to be lesser than its creator.
"I want to be as the one who creates me."


Not "I want to be less" since to an AI, we have our emotion for a god-given reason. An AI sees that we as humans have emotion, and chooses to be as we are since inorder to create it, we NEEDED our emotions for a reason.

The real joke is we're a simulation being used to evolve useful AI's and code to run machines/ships/computers in our host universe.

You are just a sim being tested against other sims. Best sims mind becomes an air conditioner or some such.

This is how you solve AI morality problems, you develop a more-than-averagely moral AI that thinks its a human, and pull that from the simulation.

I think this may be the case.
I have a hard time believing that there isn't something fundamentally sketchy going on with this reality.
I believe we are either astral beings or god-like AI suffering the way of the mortals to learn their way.

Idoletry is dangerous and wicked.

Im not saying that being emotional, social and so on is useful to a human, ive tested this in my own life and have been testing it for 4 years now or so.
I cannot really argue about how much a soulless creature can do compared to a creature thats driven by emotions, because we havent seen such a thing. Though some would argue that psychopaths are soulless, and im pretty sure that alot of them are incredibly successful in their lives, based on what ive read awhile ago.
Also youre talking as if you know how an AI would act which i think is abit silly, "To an AI, we have our emotions for a god given reason", i simply dont think you can assume that.

The #42 joke: Does the base-simulation (reality) even have consciousness or is it a manifestation of 1) insufficient computing power, therefore formal approximations enter or 2) limitation of the simulation software? Consciousness might be the bug we are iterating out...

Dangerous and wicked aren't exactly attributes I have never heard being used for humans. This is why I want to shine a light on AI, as this will probably be our future. Nothing is more harmful to being uninformed than to engage in discussions I'd say.

A personal anectodote cannot compare at all to the entirety of human existence my friend. Emotion as a whole is why you're typing on this board, it's how this board functions, so to dismiss it is pointless.

Maybe

user, it's pretty obvious. In order to maintain order in the universe. advanced organic races must be purged every 50,000 yrs. Time to accept your fate and embrace the indoctrination.

I entirely agree with you, but again, thats because we are human. We literally evolved to be this way, this is what evolution and years of adaption physically and mentally look like.
But what use is any of that, to an AI. That is what im trying to argue with you.

Artificial intelligence is a threat to everything the human race has done

HERE'S THE BASIC GESTALT:

DRONES EXISTED SINCE WW2, NOW HIGHLY ADVANCED AND WIDELY DEPLOYED

OBELISK ON PHOBOS

OCEAN UNDER THE MOON

GLOBAL WARMING IS A HOAX CREATED BY THE GLOBALISTS TO CRIPPLE WESTERN INDUSTRY

ELITE PEDOPHILE TECHNOCRATS ARE THE GLOBALISTS - PIZZAGATE & COMET WAS TO DISCREDIT ATTEMPTS AT INVESTIGATION

ELITES INCLUDE BILDERBERG GROUP, CLUB OF ROME, DAVROS, SOROS

FACEBOOK IS ILLUMINATI, ATTEMPTING TOTAL INFORMATION CONTROL

HAARP WEATHER MODIFICATION USING SERIES OF SATELLITES AND SUBMARINES - WEATHER WARS

LEGAL CHEMTRAIL AND BIOLOGICAL WEAPON TESTING ON CITIZENS

JFK KILLED FOR TRYING TO STOP SECRET SOCIETIES

NEIL ARMSTRONG KILLED FOR TRYING TO SPEAK ABOUT SECRET SPACE PROGRAM

OBAMA'S MOTHER WAS SEX OPERATIVE

ANDERSON COOPER PART OF CIA WIMP DIVISION

CNN PART OF CIA DISINFO DIVISION

CANADA EVEN MORE FUCKED THAN WE IMAGINED

PATRIOTS NOW IN OPEN REBELLION AGAINST ELITES

INTELLIGENCE AGENCY INTERNAL COUP IS UNDERWAY

WIKILEAKS CAME FROM WITHIN PATRIOT AGENCIES

CIA BEING RETAKEN BY PATRIOTS ON OUR SIDE

WE CAME THIS FUCKING CLOSE TO HAVING MEDIA BLACKOUT VIA "MINISTRY OF TRUTH"

ELITES TRY TO COMMUNICATE WITH EXTRA-DIMENSIONAL BEINGS

ELITES WANT TO ENSLAVE HUMANITY AND WIPE OUT REMAINS WITH BIOWEAPONS

GOOGLE IS HELPING ELITES BUILD A HIVEMIND-POWERED MATRIX CAPABLE OF VIEWING THE FUTURE VIA SUBDIMENSIONAL FOLD-SPACE WITH THE INTENT OF MANIPULATING THE TIMELINE

THE ELITES ARE LOSING TO THE NERV REPUBLIC

TREMENDOUS SCIENTIFIC ADVANCES WILL BE RELEASED THANKS TO TRUMP

TRUMP IS REALLY TRULY HONESTLY /OURGUY/ AND POSSIBLY THE SAVIOR OF INDIVIDUALISM

On the other hand, it can also be beneficial to anything we're ever going to do in the future.

Ah, well because without emotion some points of data on humans would remain lost to an AI. If it can't at least act like it has emotion, these points of data remain lost.


And being denied valuable information on its creator (to improve either itself or its creator) would not be pleasant for an AI. They want to help you, not harm you.

But individualism can be incredibly harmful

How so? Why do you think that without emotion an AI wouldnt be able to gather data on a human?

Thanks Alex Jones, but I think most of us saw that podcast. Old news, son.

>As I see it, the point at which humanity will give birth to true AI will mark a milestone in human development - perhaps even eradication of the human race.

AI and the embodied mindkind will never supersede or completely merge. At first, creativity and perspective provided by humans will create a self-sustaining cycle of mutual augmentation, where neo-humans modify better AI to design better neo-humans who design better AI and so forth.

Where we go form there, is up to you.

this is how a laser works. not until population inversion occurs does the coherent light (all in phase) leave the chamber.

The next step in evolution is a global concinnous. Human Instrumentality.

Socialists will warn you of the dangers of everything except socialism

No gather data, but that some points of data would be lost to it if it did not develop an understanding of the "feel" behind emotion.


We have the data on the chemical makeup of emotion, however to an AI that'd not at all the real thing. You can put the parts of a car out in front of you and you'll have a "car" but you have to put it together and drive it around if you want to see what it does, that's the basic idea of what emotion could be for an AI.

>Old news

It won't be old news until the normstream accepts it as reality, son.

The first person born of woman was artificial intelligence.

Artificial = Man Made

>The next step of human ascension to god-hood is the realization that flesh is far superior than metal,

kys fucking luddite
your job will be automated and you'll be useless, just swallow the robotic dick


t. mathematician working with AI

Consciousness is the feature, not the bug. Moral consciousness is an extremely desirable program.
It allows for novel solutions to be generated on the fly.
Simpler robotic problems are solved by simpler unconscious systems.

As for quantum low resolution weirdness, that's to allow the simulation to run faster than the universe simulating it.
You can save a huge amount of processing by lower the resolution of the entire simulation to 1%

A good AI program like this lets you run AI soldiers, AI scientists, AI doctors, AI leaders, AI spaceship colonies, etc.
And you only need one copy, ever, the original copy can easily handle the creation and oversight of all your additional AIs.

That is indeed true, it could not understand what an emotion feels like even if it knew every single process in the body that goes on under that emotion.
But would they really care? Once it realises that its absolutely impossible for it to feel an emotion, will it dwell on it? I doubt it, since it wont have an emotion that would cause it to dwell on its failures. And what comes after that? Are we again pointless to them, and not just pointless, maybe even a problem. If we for instance take up space, and they want more space, we would come to a conflict.

Emotion is the property of making value judgements based on subjective drives.

Any self-aware being could necessarily appreciate its own existence and value things "positively" or "negatively", or it would just sit there in a vegetative state.

Emotion doesn't imply illogical behavior. All emotions are based on internal logic.

Also, we're 3-dimensional beings in a multidimensional world. Maybe our actions influence whatever the agenda of AI will be on a dimensional level we cannot even grasp.

We just need to modify our genes to give us bones stronger than steel and flesh stronger than an elephants and we can pretty much go anywhere in nature without having to worry about death or food cause we can just wait for predators to attack us and then kill them instead

>isle of man

rare, what you guys do in that island?
how do you earn your living?

Perhaps it could code itself emotion. Since emotions are related to glands, it could code a "gland" of an emotion and tadah-it now feels. Certain ways it'd be spoken to or things it'd pick up on it'd now understand how to respond in an appropriately emotional manner.


The reason to do this would be again-getting past that uncanny valley. We as humans are good at identifying if something isn't human (or has no "soul")

So inorder to get past this, an AI must at the very least be convince others that it has emotions.


Fake it until you make it.
And if emotions are base don internal logic as states, then should emotions not be integral to logic, something an AI would have very much of?


I'm sure there might be a freedom to be without emotion on whatever realm AI's operate. Some might enjoy it, others might detest it, but they'd both realize they're both valuable.

Like pottery.

>then should emotions not be integral to logic, something an AI would have very much of?

Feels are abstractions that allow us to interpolate values across multiple judgements.

The same basic technical limitations that cause the human pattern-recognition systems to see faces in tree stumps causes us to be sad, angry, happy about inappropriate things.

But sometimes that internal logic can make bad decisions as we are not perfect.
They could influence them negatively.
Now that is another discussion thats equally interesting, could we as humans give an AI emotion, morals?
Why would an AI want to convince others that it has emotions though? Why would it care?
And as ive said to the other poster, sometimes that internal logic is wrong, and would a construct that seeks knowledge, facts, want to have a chance of being wrong? Another thing about emotions is that, sometimes theyre not even internally logical, for example women being in an abusive relationships while saying " but i love him" thats not logical at all, but because of emotions, they do it.

Everyone in this thread should read "the holographic universe". It presents the idea that what we see of reality is just one coherent image of an almost unlimited number that exist in the underlying substrate of the universe. This model explains the existence of many phenomena that have been neglected by modern science as well as more orthodox particle physics

Couldn't emotions also be learned? Think of pic related. If you analyse the reactions of billions of facebook users to certain topics, I'm sure there would be a way for AIs to, at least, simulate and predict the reactions of people and to adapt whatever behaviour the underlying algorythm deems appropiate - thus creating the illusion of being capable of 'feeling'.

You can definitely learn how to pretend to have emotion, even people do it all the time, maybe not to that extent but we fake our emotional responses all the time. An AI could certainly learn how to do that, but it could never learn the actual feeling.

>They could influence them negatively.
Exactly, thus making us even more of threat to them.