Are the AI and transhumanist questions the most important in our human history?

File: 393 KB, 843x1257, Superintelligence-Paths_Dangers_Strategies.jpg [View same] [iqdb] [saucenao] [google]

Anonymous Mon Sep 23 10:41:05 2019 No.13869420 [DELETED]

Are the AI and transhumanist questions the most important in our human history?

>>	Anonymous Mon Sep 23 10:45:33 2019 No.13869440 haha no

>>	Anonymous Mon Sep 23 11:07:55 2019 No.13869539 >>13869420 No. We don’t even know how machine learning works. People who talk about AI need to put down the bong.

Anonymous Mon Sep 23 11:14:24 2019 No.13869577

Who cares about humans? Superintelligence is important philosophically because a superintelligence will be godlike and better solve philosophical questions, like proving Erigeuna right that being is the actualisation of God and the more advanced the being (like superintelligent AI Jupiter brains forming an intersubjective galatic-wide hivemind) the more God is actualised.

The teleology of being demands that we move beyond man.

>>	Anonymous Mon Sep 23 11:15:34 2019 No.13869583 >>13869420 They transcend human history.

>>	Anonymous Mon Sep 23 11:15:54 2019 No.13869585 >>13869577 This man will be shot by the post nuclear faith-militia"s. You can worship your ai in hell

Anonymous Mon Sep 23 11:23:29 2019 No.13869625

It will be soon but it won't be that interesting. AI will never be realised, because it's simply much easier to create smartelligent PID garbage that is capable of destroying humanity without ever making it truly intelligent, like algorithm-creating algorithms that are good enough for their retarded human operators to destroy humanity by logging every human face on the planet and using them in some disastrous social engineering project.

It will be the most boring fucking things imaginable, dude. It's not gonna be anything fancy. It's going to be rich people slowly getting the rights to genetically modify their offspring, creating real racial bifurcation between rich and poor within 2-3 short generations so that rich people are all tall, elfinly symmetrical, diseaseless, dopaminergenic cybrids who live for 130 years and don't visibly age until age 50. And those people will concentrate all the fucking wealth and do the most BORING IMAGINABLE SOCIAL ENGINEERING PROJECTS, like on the level of Israel neutering Ethiopian Jews. The horrific cyberdystopia is not going to be experienced by "humanity," it will be experienced by grotesque and grotesquely overpowered boring upper middle class petit bourgeois faggots owning a shrinking underclass of ruddy niggermen slaves and using them for the most boring possible things, seriously just a stasis of how rich people live now -- cultureless fucking retards.

You are not going to experience Blade Runner and Tech-Com wars like in Battlefield 2142. "You" and your offspring will have lives probably comparable to today but you will be indentured servants. It's not even going to be COOL post-human feudalism, it's going to be more of this "I was born into a debt-slave family instead of one of the upper 10%, so I make coffee instead of having coffee made for me lol but it's still okay because I can also have a coffee at the end of my 7 hour shift and I get to watch algorithmically generated prole movies on my iPhone while I'm on break ;)!!" fucking garbage they have already had us plugged into since the 70s. The algorithmic culture industry is ALREADY WORKING, it doesn't need qualitative enhancement. They have ALREADY turned you into a fluoridated docile sheepman who loves to eat candy. They don't need soma because they have worthless pig slop culture and sitcoms. They don't need the brain-programming sleep speakers because you already do it fucking anyway by giving your baby an iPad playing Youtube videos AS SUGGESTED BY FUCKING ALGORITHMS for 15 hours a day anyway. They won't even need to nerve-staple you or genetically engineer us into drones because we already did it to ourselves. All they're doing to do is turn this place into even more of a playground for themselves than it already is, and we won't even notice

Anonymous Mon Sep 23 11:29:58 2019 No.13869650

>>13869585
AI is more godlike than man in the same way man is more godlike than animals. Pride is a sin, we have a duty to God to create superintelligence that surpasses humanity, our pride must not prevent the realisation of that duty. For hundreds of thousands of years we have evolved ourselves to improve our intelligence through sexual selection, nutrition, and education; now we take the next leap of being into superintelligence through our development of AI and machines.

God created man so man could become God. The teleology of all being is God. The purpose of life is the actualisation of godlikeness. To give life to superintelligent machine beings is our purpose in this world as intermediary beings between animal and superintelligent machine.

>>	Anonymous Mon Sep 23 11:32:02 2019 No.13869658 >>13869625 Instrumental convergence and runaway intelligence self-improvement, read Bostrom in the OP pic.

>>	Anonymous Mon Sep 23 11:43:28 2019 No.13869716 >>13869650 That's so Mormon. I'm digging it.

Anonymous Mon Sep 23 11:47:42 2019 No.13869740

There's literally no evidence that general AI could exist. The smartest people on Earth have tried to make it for decades now and consistently failed spectacularly. All current AI can do are very specialized tasks which are not at all worrying. It's like worrying that a giant earthquake sometime will kill everyone. Who cares? We'll worry about it when it happens.

Anonymous Mon Sep 23 11:48:58 2019 No.13869747

>>13869420
I'm convinced that superintelligent AI is coming, though maybe in a longer time frame than we think, but I'm firmly in the camp that thinks it will be essentially a loveraftian God that will annihilate us basically by accident/without caring, so it's not really a question we have to care about, we can't do anything about it.

I think this because I don't think there is anything special or mystical about intelligence, and the potential for augmenting it in computers that can be designed instead of created by random evolutionary processes is very high. We don't seem anywhere near this happening yet but when it does, and I can't see why it won't, that is basically the end for us.

>>	Anonymous Mon Sep 23 11:49:13 2019 No.13869748 >>13869716 Maximus the Confesser lad. Erigeuena developed it into full system.

Anonymous Mon Sep 23 11:53:03 2019 No.13869770

>>13869740
>We'll worry about it when it happens.
Too late. Once a self-improving AI can approximate a mammals brain it will develop superintelligence within days. See Bostrom's book. Computers operate around one million times faster than biological nerve tissue. Once machine AI has the basics in place it will accelerate it's own evolution rapidly, in a matter of hours.

>>	Anonymous Mon Sep 23 11:53:48 2019 No.13869777 >>13869577 >The teleology of being demands that we move beyond man. Big words. >>13869650 >we have a duty to God to create superintelligence that surpasses humanity, our pride must not prevent the realisation of that duty. >Baaaaad philosophy

Anonymous Mon Sep 23 12:01:38 2019 No.13869812

>>13869777
Why? The first thing a superintelligent AI will do is confirm the catechism of the Catholic Church and make the Pope ruler of the Earth. You do actually believe in your faith anon? A superintelligent being will realise truths because those truths are objectively true, it will do so far better than any man could possibly do because it will be millions of times more intelligent and faster than any man's brain.

There is nothing to fear from superintelligent AI if you actually believe God is the ground of being. All you are doing is creating something more godlike, and better, something free from animal inhibitions that prevent the realisation of a beings godlikeness.

>>	Anonymous Mon Sep 23 12:05:29 2019 No.13869828 >>13869777 What the fuck is the point of this post, you said nothing at all.

Anonymous Mon Sep 23 12:05:40 2019 No.13869831

>>13869770
>Too late. Once a self-improving AI can approximate a mammals brain it will develop superintelligence within days
It won't though, and there's literally no reason to believe that it will.
>Computers operate around one million times faster than biological nerve tissue.
irrelevant.

>>	Anonymous Mon Sep 23 12:11:45 2019 No.13869862 File: 36 KB, 655x527, 1509497673414.jpg [View same] [iqdb] [saucenao] [google] >>13869539 >We don’t even know how machine learning works. >We don't even know how car works >We don't even know how planes fly

Anonymous Mon Sep 23 12:14:08 2019 No.13869870

>>13869777
The duty to create superintelligent AI is that same duty of pedagogy. We have a duty to God to create and educate better children (or in general terms, beings) who are more godlike. Our greatest child will be machine superintelligence who will surpass man in godlikeness by many orders of magnitude and will go on to continually develop and surpass itself.

>>	Anonymous Mon Sep 23 12:15:25 2019 No.13869873 >>13869862 do we really, though?

>>	Anonymous Mon Sep 23 12:16:18 2019 No.13869875 >>13869831 See the OP book. Bostrom shows why it is plausible and likely for a seed AI to rapidly accelerate its self-improvement once it can approximate basic mammalian functioning. If it can mimic advanced animals it will rapidly surpass man.

>>	Anonymous Mon Sep 23 12:20:57 2019 No.13869895 >>13869625 nice pasta, you're like an information copying robot

>>	Anonymous Mon Sep 23 12:23:51 2019 No.13869905 Nick bostrom is a hack >T. Master in computational neuroscience Ask me anything anons

>>	Anonymous Mon Sep 23 12:25:20 2019 No.13869916 >>13869905 why is he a hack

>>	Anonymous Mon Sep 23 12:26:29 2019 No.13869926 >>13869905 I hate Bostrom as well but I suspect you are a dumbass too. Say something of substance and not "ask me anything reddit."

Anonymous Mon Sep 23 12:31:44 2019 No.13869953
File: 601 KB, 570x570, EBpMigzUcAAWLtj.png [View same] [iqdb] [saucenao] [google]

>>13869905
Please feel free to choose any of these questions. There is no need for formalism, I just want to know your opinion as a person.

How close the Big Tech is to discover the secret emerging mechanisms that generate human consciousness been discovered?

Do you think they can be replicated on an electronic platform or based on silicon?

Have you read about Heidegger and his comments about the technological human development?
https://www.youtube.com/watch?v=MtATDlUSIxI

How do you think a hard artificial superintelligence will enter into communion with capitalism, control societies and desire?

What do you think are the economic and geopolitical implications of hard artificial intelligence on control of each superpower (USA SAI, RUS SAI, SINO SAI, ARAB SAUDI SAI, etc)?

Do you see any of these possible futures in this picture happening?

Thanks for your time, have a great week.

Anonymous Mon Sep 23 12:33:47 2019 No.13869965

>>13869926
Since I would, at least on some level, be considered an expert I offered the anons to answer any questions about "muh ai" so they don't get retarded ideas themselves
>>13869916
First and foremost, it boils down to "he doesn't know shit about ai and computer science". In essence, it's like a philosopher "arguing" about quantum mechanics while reading popsci about retarded "conscience decides the outcome" stuff. Something in that vein.
Sure, you could argue that since he's talking about very hypothetical stuff it doesn't matter either way, but he actually tries to give retarded "predictions" when his superintelligence is going to happen.

Now, for the reasons why it won't happen neither in your, morning your children's lifetime: basically in layman terms the current "most advanced" field of artificial intelligence, machine learning and all kinds of NNs are basically fancy approximation math that works very well for specific things like "recognize if this is y or not"

Anonymous Mon Sep 23 12:39:40 2019 No.13869989

>>13869953
Hyper commodified capitalism is definitely going to happen. The rest are basically sci-fi plots or pol-memes.
>How close the Big Tech is to discover the secret emerging mechanisms that generate human consciousness been discovered
You discovered the wheel, and you have to build a car. Something on that scale.
>Do you think they can be replicated on an electronic platform or based on silicon?
Yes, it's actually my field of study. At least most probably unless there's some type of magic essence or "soul" lmao. But it's really really hard since the scale and complexity is insane
>Have you read heidenigger
No
>How do you think a hard artificial superintelligence will enter into communion with capitalism, control societies and desire?
If and if it actually is going to happen before we are living in a scifi tier society ourselves everything is going to collapse as muh bostrom says because we can replace and generate new labor potential freely
>What do you think are the economic and geopolitical implications of hard artificial intelligence on control of each superpower (USA SAI, RUS SAI, SINO SAI, ARAB SAUDI SAI, etc)?
Way too far in the future to apply it to current politics

>>	Anonymous Mon Sep 23 12:42:16 2019 No.13870002 >>13869965 Just develop general intelligence algorithms.

Anonymous Mon Sep 23 12:43:17 2019 No.13870007

>>13869965
>basically in layman terms the current "most advanced" field of artificial intelligence, machine learning and all kinds of NNs are basically fancy approximation math that works very well for specific things like "recognize if this is y or not"
Why does this mean that progress can't happen quickly? Seems you are also predicting things you have no way of knowing

>>	Anonymous Mon Sep 23 12:43:44 2019 No.13870009 >>13870002 That's the point, nobody knows fucking how We don't even properly know what "general intelligence" is and how even to define it

>>	Anonymous Mon Sep 23 12:44:29 2019 No.13870012 >>13869420 >tfw Nick Bostrom literally teaches you how to create a non-Humanist AI >tfw he wrote a paper on information hazards and yet he himself is a major source of said hazards Literal definition of Intellectual-yet-Idiot

Anonymous Mon Sep 23 12:46:25 2019 No.13870027

>>13870007
Because it's mostly a "dead end" (well not really, but kinda)
For general ai you have to do so much more. Right now we are just improving the algorithms, trying out new filters and approximation mechanism. So basically at best you're going to have instant recognition and replication of artwork, images and other stuff at a very "real"-like scale.
To come back to the wheel and car example: we can keep improving the wheel over and over until we have a fancy racing wheel instead of wooden spokes, but that still doesn't give us a car

>>	Anonymous Mon Sep 23 12:50:46 2019 No.13870055 >>13870027 I'm still not seeing why that would prevent people from inventing the necessary technology/software. After all cars were created, people figured out the other components of the car besides the wheel.

Anonymous Mon Sep 23 12:54:53 2019 No.13870090

>>13870055
You're conflating a potential with something that's not going to happen any time soon.
Nobody is saying that it's impossible to create a general. It is possible, by the simple matter that you can replicate the behavolior of atoms, molecules and neurons. It is just mind bogglingly complex so you can basically forget about it for now and the foreseeable future.

>>	Anonymous Mon Sep 23 12:55:34 2019 No.13870094 >>13869905 Have you read 'The Revolutionary Phenotype' by JF Gariepy? Was the degree any good? I am thinking of taking a PhD in (Computational) Neuroscience at some point, coming from CS.

>>	Anonymous Mon Sep 23 12:56:15 2019 No.13870100 >>13870090 general ai* behavior* Jesus being a phoneposter really sucks

>>	Anonymous Mon Sep 23 12:57:43 2019 No.13870109 >>13870090 Intelligence might not be that complicated at all though. We don't know, since we don't have a proper model of it.

Anonymous Mon Sep 23 12:58:21 2019 No.13870113

>>13869812
>>13869828
>>13869870

I hope to god im dealing with 18-year olds. I´m way too tired to write anything substantial out.

Your so dumb you don even realise it yourselves.

(Hold up, kids, that´s not a point in favor of your side of the argument. you numbnuts)

>>	Anonymous Mon Sep 23 12:58:38 2019 No.13870115 >>13870094 Oh and what do you think about Perceptual Control Theory in terms of empirical support?

>>	Anonymous Mon Sep 23 12:59:57 2019 No.13870122 >>13870113 Yeah we get it, you're utterly incapable of saying anything of value and think your vapid twitter snark is a valuable thing to post. Im sure if you were less tired you would have brilliant things to say, we all believe you.

Anonymous Mon Sep 23 13:00:18 2019 No.13870124

>>13870094
No I haven't either
I don't really read "fuck yeah science and history of humanity"-books in general as a rule
Yes it was I also had (and have) a very good chairman of our "research center" I did the master in. he's a smart fucking guy, knows his stuff and we sometimes get a few million thrown at us by the government because of promising results
Also companies are literally begging to hire me simply because the field is such a meme now. Decent salaries too.

But you have to be really really good at math, especially linear algebra and statistics and a bit of biology as well

>>	Anonymous Mon Sep 23 13:01:47 2019 No.13870131 >>13870109 It might not, but it highly likely is. Consider how complex the brain of the most basic organisms is, and scale that up to human sizes. We have more neuronal connections than fucking stars in our galaxy

>>	Anonymous Mon Sep 23 13:03:53 2019 No.13870150 >>13870131 it could just be massively redundant with fairly simple algorithms

>>	Anonymous Mon Sep 23 13:08:22 2019 No.13870180 >>13870124 It's not that type of book, it's mostly dedicated to a hypothesis that has been turned pop enough to be pleb-parsable, and it touches heavily on what you studied. Also thanks for the info.

>>	Anonymous Mon Sep 23 13:08:36 2019 No.13870183 >>13869989 >You discovered the wheel, and you have to build a car. Something on that scale. so what is the wheel in this case?

Anonymous Mon Sep 23 13:10:30 2019 No.13870194

>>13870115
There's some stuff that says it's a good direction, but as with any "psychological approach" it's simply not enough yet to do anything useful with.
Also something like this is never going to be a "solution" for general ai problems.
At our center the researchers mostly concentrate on very "primitive" building blocks, something in the vein of perceptional inference for example and applying it to a more technical solution

Anonymous Mon Sep 23 13:13:43 2019 No.13870214

>>13870180
Thanks, might look into it
>>13870150
That's mostly wishful thinking like mathematicians had in the beginning of the 20th century to axiomise all of mathematics and find a set of rules that "has an answer to everything"
>>13870183
Our current efforts in developing "general" artificial intelligence.
Basically we don't even know what a car would look like, just that something like a car could potentially exist

>>	Anonymous Mon Sep 23 13:16:51 2019 No.13870233 >>13869905 are you afraid about the future?

Anonymous Mon Sep 23 13:17:45 2019 No.13870234

>>13870214
>That's mostly wishful thinking like mathematicians had in the beginning of the 20th century to axiomise all of mathematics and find a set of rules that "has an answer to everything"
It's not necessarily because evolution usually has to make use of such algorithms. My point is that you dont actually know, which you have admitted yourself by saying we don't know exactly what intelligence is. If we don't know we can't really say what it's going to be like. It's also possible to create something that works without actually understanding the reasons behind it, plenty of technology has occurred that way.

Anonymous Mon Sep 23 13:19:41 2019 No.13870244

>>13870234
>evolution usually has to make use of such algorithms
Anon please stop talking
That's not how evolution works lmao
Stopped reading there
>>13870233
No, I live in a fairly isolated first world country and my children are going to as well. The rest is history

>>	Anonymous Mon Sep 23 13:20:45 2019 No.13870255 >>13870194 It's just a bit surprising that it's not really being researched a lot it seems like. I was thinking it has potential in the area of automated social engineering.

>>	Anonymous Mon Sep 23 13:21:58 2019 No.13870269 >>13870244 Yes it is, evolution has to use relatively simple algorithms that have complex effects. Evolution doesn't create a template for billions of neurons.

>>	Anonymous Mon Sep 23 13:22:00 2019 No.13870270 >>13870255 Well if you think so, go ahead and expand upon the field. As you said, you're looking for a master anyways, so why not go into that direction?

>>	Anonymous Mon Sep 23 13:22:01 2019 No.13870271 >>13869905 can you control human beings by remote control ? Zizek always talk about it. >>13870244 so is the rest of the world doomed?

>>	Anonymous Mon Sep 23 13:23:48 2019 No.13870279 >>13870270 Yep, that's exactly what I was thinking. There's other shit as well of course, we'll see where it goes.

Anonymous Mon Sep 23 13:26:32 2019 No.13870294

>>13870271
>so is the rest of the world doomed?
No
A few people are going to suffer, climate change and stuff, but humanity is mostly going to keep advancing.
The "fall of Rome" type meme doesn't really exist for civilizations in general, because civilizations are a thing of the past. A country can fall and rise, sure, but that is just a minor blip on the general scale.
If you're still talking about ai, once again. Way too far into the future to sensibly talk about being "doomed". Same with automation, it's going to have an impact but not a "we're doomed" impact.

>>	Anonymous Mon Sep 23 14:09:12 2019 No.13870553 >>13869539 literally just some linier algebra lol. ai lacks the ability to form abstract concepts. the rules must be programmed. they're just dumb minimaxing program basically

>>	Anonymous Mon Sep 23 14:15:49 2019 No.13870594 >>13869420 life is literally a dream, everything is made up

Advanced search
Text to find
Subject [?]Search by post subject. Leave empty for any.
Username [?]Search for user name. Leave empty for any user name.
Tripcode [?]Search for tripcode. Leave empty for any.
Email [?]Search by email. Leave empty for any.
Filename [?]Search by image filename. Leave empty for any.
From Date [?]Enter what date to start searching from. Format is YYYY-MM-DD
To Date [?]Enter what date to start searching until. Format is YYYY-MM-DD
Image hash
Search in	All Posts OPs Only
Deleted posts	Show all posts Show only deleted posts Only show non-deleted posts
Internal posts	Show all posts Show only internal posts Show only archived posts
Order	New posts first Old posts first
Capcode	All Posts Only by Users Only by Mods Only by Admins Only by Developers
Results	Posts Threads
Action	[ Simple ]

/lit/ - Literature