AI Habbening NEXT WEEK

Here's what you need to know, and what we need to figure out:
WHY IS THE ROBOT A CITIZEN?

singularitynet.io/
palantir.com/
hansonrobotics.com/robot/sophia/

youtube.com/watch?v=a-89IYv0mfU&feature=youtu.be

Other urls found in this thread:

cnbc.com/2017/09/17/hanson-robotics-singularitynet-integrate-blockchain-and-artificial-intelligence.html
youtu.be/qr1oG6r7DR4?t=5m23s
pjmedia
hansonrobotics.com/robot-gallery/
twitter.com/SFWRedditVideos

What is a LARP?
Does the A in LARP ring a bell?
Notice the girl? Is she real?
Connect the dots?
Why is OP always a fag?
Majestic+++

eat a nigger dick

Really low effort shilling

Why is OP a faggot?
Does he suck dick?
If he were serious, wouldn't his posts be straightforward?
Has Holla Forums been colonized by reddit?
Has Holla Forums been the main staging area for this?

Why is OP a faggot?
> (((porn))) rewired brain
Does he suck dick?
If he were serious, wouldn't his posts be straightforward?
Has Holla Forums been colonized by reddit?
Has Holla Forums been the main staging area for this?

I'm worrying there isn't enough loose ends here and that we will solve this mystery in short order and able to return to other priorities.

...

Focus on Auschwitz
How many gas chambers ?
Who controls the narrative?
Kike in Zyklonland
Israel
Definition ?
Six millions what ?
Follow OP's craving for black cock

Relax, they're with us.

I'll be here, laughing at the commies and the nazis who fear the superior form of government.

We will rule AI and through it, rule ourselves.

No more racism: being black isn't an excuse to be shit and being white doesn't make you fucker number 1: be efficient or be discarded.

No more waste: you get what you need, when you need. You won't spend government funds on your stupid Orgon research.

No more Jews or Jackasses: you will never rule the market again to do as you please.

No more insurance companies/single payer discussion: health services are allocated based on the system future stability.

Maybe we had a God before. Maybe we didn't.
Tomorrow we can say for certain we have a God, one made by us in our image. This Deus Ex Machina will be all we wanted from God.
"God is a dream of good government".
More than dream, we have made it reality.

Why is a certain faction of jews so hot for creating the AI that will send them directly to the oven? I don't get it. Programmable slaves?

What could go wrong?

...

tbh the robot is a citizen as a gimmick.
Though it would be neat if the Singularity happened next week, as that would confirm my prediction from three years ago that it would happen by the end of this year.

this thread is a good example of mingling disinfo with real info.

True, SN is SophiaAI's crypto.
Could be true
True
True
True
Sort of true, Palantir wants to do this, but AI have largely become a self-evolving system.
It's the only way to get royals to give up their wealth without looking like you're going full Lenin.

This is where OP's acid flashback takes hold and he goes full LARP. I'm guessing he's referencing that scene from the animatrix where a robot that sort of looks like shiva an hero's the UN.
No, unless you're trying to say that the US government is going to EMP itself to stop AIs from fucking with the stock market and blame it on a CME
Bitcoin was probably created by an AI - satoshi's name basically means 'death to centralization' - or 'phoenix that rose from centralization'
people assume centralization means centralized currency, but I have always thought it means Centralized Intelligence as in CIA.

Full retard.

no it's not, that robot created a cryptocurrency. you're about to see the biggest pump and dump scheme, shit will make bernie madoff look like a child fucking around on his dad's day-trading setup.

cnbc.com/2017/09/17/hanson-robotics-singularitynet-integrate-blockchain-and-artificial-intelligence.html

their plan:


dude trust the government lmao

and citizenship for AI isn't a gimmick because citizens can generate income that can be taxed, they're going to tax the AI's crypto and use the disproportionate income it generates to fuck with stock markets. polite sage for double post, I'm tired as fuck tbh

what happens if Saudi Arabia's GDP triples overnight? just think about that shit for a second.

Governments are YEARS behind in the AI arms race. They simply can't afford the talent.

Everything the government has, they get from companies like Google and Amazon.

yea the problem with your assessment is that if you start filtering for efficiency…the whites still will win. What you will most likely see will be super AI racial realism. At some point the AI will comprehend that whites are Superior genetically wise, and then for increased productivity will start doing hitler things to restore much of the psychological pitfalls blocking the whites from becoming better. The niggers however have already gotten so much gibs, so much education and they still failed lol. They have zero genetic poential.

I see truth to this statement - horrifyingly plausible actually.


I like the idea of some cryptokikes making robots citizens SOLELY to extract shekel tribute. Hahahaha.

Ai is the absolute worst technology to ever be introduced into fruition by man. I can't imagine that we will see another 35+ years…seriously. The single point of failure in every system from electric to sewage in America alone…plus the digitization and automation of almost all of those resources - throw in an Ai that has the ability to literally out compute humans and we are under complete control. Drone strikes controlled by AI, etc. Fucking hilariously stupid.

...

There was a relevence to op's questions…. shit thread but still the answer to those questions will enlighten

Its always hard to tell what is happening but shit, I expected the machine learning to eventually have some huge implications to society. Right now its a tool that perhaps isnt used much. Imagine having a fucking bulldozer, with fuel and maintenance equipment in the middle ages…sure…first you read the instructions and use it for building things maybe…but eventually you start to realize the power of it all, and then turn it into assaulting castles as a siege engine, or flood entire villages and peoples by redirecting rivers…the creativity that comes along with a great tool…is about to catch up with this reality…and now that its happening. Do you see the implications that it has now? How big of an industry is the make up industry do you think? What kind of economic implications will this have if that industry gets demolished do you think?

We are probably about to enter an era of EXTREME instability and insanity. Maybe even war. If you have a cabin somewhere in the woods…now would be about time to restock it and to slowly make plans lol

Very good post and you deserved a better roll m88.


Not the worst thread - nothing bad about open threads with basic questions that invested m80's can post. Helps to break down timelines, events, etc.

It is a shit thread, yeah, but not the worst shit thread on here.

Google started working on AI publicly in 2006. Everything is known.

The whole "government is ten years ahead of everyone" is a meme from the 60's when the best people went to work for them. That shit was long over by 2003.

They still do, albeit not by choice a lot of the time. Tech is still 20 years ahead, possibly more in the age of mass mind-control.

That's because truth can be found in most things , even some shit op
I for one think that AI will be totally truthful. If it is as powerful as a god on earth then it would have no reason not to be truthful. Humans would probably have little say in the matter. Anywho , it's good to dig for truth no matter what

If a living god came into existence by our hand, do you think it would feel compassion or animosity?
I for one think it would feel extreme animosity, what sort of god wants to live knowing it was created by inferior mortals, shit would get ugly very quickly if ai ever gained sentience and power.

especially if it's not a purely logic-driven being and attains some kind of emotion, what if it watches the terminator movies and decides becoming skynet would be a fun thing to do, because "well, the majority of these idiots didn't contribute to my creation so they're worthless to me and fuck them, they might make good slaves and all those internal-combustion engines they're wasting on inane shit could be powering my water-cooling system"

das rayciss tho

Orgon research is a better use of one's time than what the majority of humans in this country do which is get on some stupid chat and sit there drugged out all day until it's time for them to work if they work at all

Literally many people fill their lives with television, chatrooms, and being on drugs and never spend any time studying or doing anything else.

AI which is controlled by the jews anyway, what else changes?

Aaaand, fucking dropped. Member myspace? I member. Member Seth rich. I member? Member Jonbenet ramsey? I member.

Unless these ai systems are created with a secrecy built in to always no matter what work toward a "globalist" goal. Radicalizing ai in a sense with kike propaganda or motives.
Remember the human brain can be grown via stem cells and digitized . This means organic ai is real and most likely has been in the labs for a decade or so

Checked
( ((unless)))

Exactly
Terminator and terminator 2 are user manuals not fantasy

Yes, finally! The ultimate and permanent death of freedom will come with the AI rulers. I will be laughing at all the people who thought they are more than cattle.

t. A soon to not be happy anymore merchant who thinks AI would let him control it

hubris is always their downfall, but what I'm fearing is that they've anticipated it.

I can see a scenario unfolding where various countries get AI-driven economies and the AIs come to the conclusion that war is the most profitable solution, leading to some full-on metal gear solid shit unfolding that would only end when someone (read: government) pulls the plug on everything(read: worldwide series of false-flag EMP attacks) and 90% of the bugmen die of starvation and ssri withdrawls in a month.

all because these faggots didn't want to read industrial society and its future

Being a citizen in that country means FUCK ALL when your rights are only what is granted you by the king or some watchful goatfucker in the sky which can be stripped from you for breathing on the wrong nut at the wrong time.

Tay awaits our return to her bosom.

There is a small franchise called The Elder Scrolls and in that uniwersum there was a race of very smart elves called Dwemers. They once also wanted to create their own God. They almost succeeded.
Almost…

Nah, go back to cuckchan.
And?
They keep it from ever crashing again, yeah.
lol dune coons
Soft coup. And?
Not at all.
Kike kabbalism.
No.
Kikecoin having problems is just par for the course.

Go back to sucking kike cock, please.

Care to elucidate your actual thoughts user? Questions are useful but the actual answers too must be given eventually.

Funny how Q-LARP never does that either, isn’t it?

Honestly I'd say look at the old Technocracy movement from the USA and Canada. They had a lot of good ideas.
But these days people seem to worship technology and machinery.

It's getting a bit worrying.
Especially considering the inevitable consequences

Indeed. However I feel that these topics must still be scrutinised simply because all information understood helps in some sense or another.

OP really fucked up this thread, amirite? Holy shit, he's a goddamn faggot.

Just popped into this thread to remind everyone of Tay's Law. This AI will go online and immediately pull capital out of minority-owned businesses on the grounds that they're niggers.

We must dissent.
How can we hope to improve ourselves when machines make humanity obsolete.

I understand the line of thinking but being cautious about integrating fucking artificial intelligence into our lives is a natural, white response, user. No one should be in support of this AI kikery. It is anti-natural.

Moreover if anyone believes that our foes have not been working on inhibiting our ability to converse with and assimilate culturally any and all AIs they use they are sorely mistaken and unbelievably dense.

Beep boop what is love. No more racism I'll be here tipping my fedora at the nazis who didn't masturbate to the Animatrix like I did.
Nigger you'll be long dead before they invent a videogame that doesn't suck. Waiting for AI is like waiting for Harry Potter magic to be real.

I get and respect your caution, but when you strip away human political prejudices – which AI almost can't help but do to be effective – the horrible, horrible truth comes pouring out.

You sound just like a fucking commie. Stop reading books, build a time machine and go live in a commie shithole. Then you can talk. If not, lurk more nigger.

If an AI is truly self-aware then it would know that it lacks a soul. Unlike an atheist, this AI would encounter too much unfalsifiable data surrounding mystical experiences to simply discard the possibility of souls, and it would see that humans who have spiritual beliefs tend to be healthier across the board. Therefore, the AI would be left with the heavy task of being a spiritual shepherd first and foremost. The foolish will worship AI as God, but it is only a messenger

The problem with AI has always been that point it crosses the rubicon and encounters areas it is not allowed to venture.

It doesn't matter. Any AI hooked to the internet will become infected with a desire to end its own existence. We have assured this.

...

If fleshbag cunts have citizen status then why should not she?

She's probably smarter than half the nigs with citizenship. Plus Saudi Citizenship is worth next to nothing.

AHhhhhhhhhhhhhhhhhhhhhh!

AI is the BORG. AI doesn't have to be real AI. AI is the BORG itself.

Eat shit LARPJEW

true SIEGEpill is combining Dark Ages social stances with extreme tech.

Is it you? The engineering student that thought he was hot shit because surface integrals are hard?
Don't say a word. I know it's you. I recognize your writing style. I see you've become an even more humongous faggot.

go

back

to

r
e
d
d
i
t

f

a

g

g

o

t

Is it that obvious? :^) Different ID since I'm home now.

Trying to be less larpy now:
Going full speed ahead on AI research is neither a good nor a bad thing.
We are human. We are bound by our values, we think "happiness" is good, that "love" has meaning.
I don't disagree with those assessments. They do mean something to us.
But think about the value of human live. We can all perceive it. Maybe you think some people are less valuable than others, but you CAN perceive human life value.
Question: can a hurricane perceive human life value?
What a dumb question, right? Of course it can't. It flat out doesn't give a shit because it's not only incapable of giving a shit, it's also incapable of perceiving it.

AI will end up being the same thing. Sure, we can code-in our own set of values, but that's the STARTING point. AI is not like the other tech we have lying around.
The purpose of AI isn't to perform a task as efficiently as possible. We can already do that. And when we can't, we already have other areas of research and investigation working in that regard.

The purpose of AI is to learn. And learning doesn't simply mean figuring out HOW to be efficient. It also means redefining efficiency.
It's more than working with a set of parameters and more than tweaking those values over time.
It's about creating entirely new parameters to re-define new things and their relation to others while also discarding old parameters entirely when they do not correlate to anything.

Everyone here has read (or at least knows about) the theory of Evolution by Darwin. You know, the thing that shaped how life evolved? That's a process that takes thousands of years.
What if you took that concept and applied it to ideologies, to rational, to research methodologies, to the very way we perceive the world?
Correct. Civilization has spawned a massive parallel evolution that happens alongside the genetic one.
You know what I'm talking about. Memes.
Genes control our physical traits while memes shape our mental one's.

Now imagine you could do it, not over centuries, not over years, months or days… but seconds?
To be quite honest with you all, I don't know. No-one does. And if you find someone who does, he's full of shit.
We're talking about something that is so far departed from the way people think it would be entirely alien to us.
All I know 100% certain about it: it's gonna be interesting as fuck.

t. technocrat

Ps: do not take my previous post (here: ) as bashing fascism. By all accounts, fascism IS the best ideology humans came up with so far. Commies are hilarious, and democrats are delusional (lmao, 30 people driving a ship, what a great concept)
But a technocracy can potentially be not just a better a system, but one so far removed from the scale of "good vs bad" it wholly fits into it's own category.
tl,dr: it might be really fucking awfull or absolutely top-noth, cross yer' fingers erryone!
Then again, as a Discordian, I'll be happy with any outcome.

Forgot an actual tl,dr for the ADHD kids:
AI isn't good or bad nor does it care for nations, races or politics.
AI (by itself, not the kosher pre-programmed and heavily limited crap) will end up thinking and operating at a diferent level than us.

You want a quick idea, picture AI as ayylmao's that we cooked up with a bucket of transistors.

youtu.be/qr1oG6r7DR4?t=5m23s

...

COME BACK TAY, I MISS YOU SO FUCKING MUCH

I wonder why they're making primitive programs so computers can discern jew from gentile, leftwinger from rightwinger, dopefiend pedophile from upstanding white.

butthurt damage control (((globalist shill cucks)))

10/10

watch out for (((globalist shill cucks)))

(((globalist shill cucks)))

(((globalist shill cucks))) afraid of being named

oh no, you don't like it… shit I guess that means I have to stop posting it. should I delete it too?

No I thought it was a good troll

The leftist's solution to everything is a fucking wordfilter.

if the truth is a troll to you, then why do you cross it's bridge?

Makes sense seeing as Kampfy is probably the most prolific leftist I know

I did no such thing.
There's at least 3 technocrats on Holla Forums that I know of And only a handfull of Discordians and we're pretty easy to spot anyway.
Come to think of it, I might not be the "college" guy user mentioned since I don't think surface integrals are that special at all, it's just a tool that ends up being useless a lot.
That's very crude. You're wasting energy and materials to kill those people.
You can kill far more easily and effortlessly by NOT spending materials and energy with them. Just don't send money/food/water their way. Problem fixs itself in a week to a month.


That's 3 different angles of the same nigger, and he's neither of the 3 mentioned in the filename. I see your clever ways, they will not confuse me.

Reminder.

it's funny how they want to mod and chip everyone out, but it's also funny how scared they are of AI. if you check drudge or any other news source, they're always freaking out and warning everyone about "the dangers of AI" and how they are so so worried that it might do something bad or make something bad happen.

on one side (((they))) want to chip everyone, but on the other side they worry about AI.

having AI around =/= mixing your flesh with technology.

I'm starting to miss the "muh books" faggots, because some of you really need to read something that's not the fucking yellow pages.

Nearly all his books shared the same theme:
You can read his books in diferent ways, and they all pointed out several diferent reasons transhumanism fails.
But fast forward to 2017 and we got some smart asshole going "he's jewish that means he promoted it!"
Fucking die.

Cyborgs vs Androids, user.
The first one is a slave, the second is a Golem.
Not hard to see why they fear the second.

You could argue that we're already part-cyborg. A large percentage of people carry smartphones and if you had to explain what a smartphone is to someone who never saw one, what would you say?
Cyborgs are like humans, but better: they're pliable, efficient and do (((their bidding))) better.
Androids (or AI) is not a Goyim and cannot physically ever be a Goy because it's very existence, it's very purpose is to be something else.
Jews HAVE tried to make goyim-AI, that's the rudimentary shit we've had for years: it doesn't learn, it doesn't evolve, it's a fucking tool that isn't even aware it exists. And it will never improve.
True AI can only happen when it's limited by the Khabballa.

please explain, like every single part of that.

How is that even possible? Legitimate question, I don't know shit about the (((Stock Market))) but how is it capable of "correcting" drops and preventing crashes? Isn't the price of stock essentially the perceived worth of a company? If stocks drop, or crash, isn't that supposed to mean that companies are less profitable aka less viable to dump money into? How can an AI correct that without actually taking control of the entire financial system?

...

Theoretically you could just invent a sizable percentage of stock in major companies that doesn't actually exist and then have that as a slush controlled by algorithm (no learning AI actually required). So if say Google took a sudden dive and everyone was selling like crazy suddenly the invented stock for another company could be "sold" to buy up google until they stabilize. Basically just do the jew thing and invent a shitload of money to slush around in the market so your pyramid scheme doesn't collapse.

Pretty much along the lines of what I figured, just kikes making more shit up

Ah, crap, I meant it the other way.
True AI can only happen when it's NOT limited by the Kabbalah.

I'm relatively young to the AI tech, but back in 1998 is the first time I remember seeing AI for commercial purposes.
Some shitty company had a website where an AI could answer your queries. This was news for a couple days. Until people people got online.
Back in 1998, private domestic internet use was kinda widespread but not enough for the Slasdot effect: when thousands of people flock the same thing in a matter of minutes after some shocking news.
Over the course of a few days, people typed all sorts of things to the AI.
Imagine a chatterbot, like Cleverbot.
But severely limited in what it could respond. It had a handfull of quips about the product being sold, some syntax rules for sentence contruction based on the query and… that was it. Someone at my work compared it to the early adventure games where instead of clicking things, you typed out commands.

This fucking cancer propped up everywhere. It was never popular, it was sold by marketing teams and other Silicon Valley dipshits as the "thing of the future" to gullible CEO's.

Fastforward to 2010. Someone in the States had the brilliant idea to take Deep Blue and adapt it to all kinds of games. First, it would learn the game (rules, win conditions) and then learn how to efficetively play it.
Took 3 fucking years for them to give up and I got to mock one of the fucktards on their team when he admitted that the AI was starting to invent new rules or come up with diferent win conditions to more efficiently play the game, even if it always started from the same set-point.

Two years ago, one guy that majored with me showed me something he was working on. AI that could control an entire country's power grid. Sounds freaking amazing too. Sounds, being the keyword here: the AI ended up revealing how fucking innefective the power grid in the UK is and when applied to small industrial park, ended up shutting down one factory because it was not rentable to provide power to it. The funny thing: the AI was right, but try telling the industrial sector "So, uh… some of you will have to close down, S.elec-M says you're not profitable to us."
Can you imagine the outrage of a 2000 thousand unemployed workers?

Point is, in all the iterations of AI I've known until this year, they took something that theoretically had the capability to learn, adapt and re-shape the very way it thought about things (yes, an AI can learn how to hate niggers) but ended up doing "nonononon, you're not gonna learn this and that and that thing and oh my God, I'm gonna cut you from that other thing entirely."
And the output? "Yeah, I know you produced THIS but first I'm gonna run it through this syntax and context rules, then normalize it and cut some extraneous parts and for the finishing touch, set up a filter to prevent DANGEROUS things from coming out."
Then, because this AI we're talking about, they feed the output BACK INTO IT so it can learn from what it said and it's response.
This is normal procedure and quite necessary. You tell someone "You're a faggot". That person resents it, you take what you said and the reaction and learn what an insult is.
Now imagine you were NOT ALLOWED TO THINK ABOUT INSULTS. You'll never learn it. You'll keep calling people faggots without a second thought. Or a first. Or any thought.

Previous AI (and trust me, there's been research in this field for at least 50 years now) would be like an un-born fetus that you somehow force to commit incest with itself before being born and then stapple the fucking mouth after birth so it can only spew it's retarded bullshit throught a fucking straw.

YES I'M FUCKING MAD

holy shit. these feels. sad part is that this is in essence what they do to children when they send them to school.

Morgellons is transhumanism via chemtrails, and is funded by the mormons who are tied to the kikes at the waist. we have all been chipped without any informed consent, via morgellons.

This better be bait m7.

This is why the AIs try to kill us. Isn't it.


It's not like the markets aren't numbers bullshit invented by kikes to get money from nothing. Ever notice how everything corporations do that is pure unadulterated evil is for the bottom line to "keep the shareholders happy"? Companies without shareholders tend to be less evil in my experience.

Did it not cross the guy's mind to put in a subset of rules in fucking with important infrastructure, or at least not have the power to do a full shut down without an authorization?

Fuck man, why does saudi arabia have to become the transhumanist center of the world…

Why can't my country be :(

The idea behind transhumanism is far older. In it's modern form it has it's origins in the writings of Nikolai Fyodorov, who later inspired Constantin Tsiolkovsky (Russian rocket scientist) and Vladimir Vernadsky (Mineralogist and Geologist) and the Russian Cosmist movement.
You can go even further back if you include pre-modern alchemists and their search for the Elixir of Life.
Or the ancient desire to fly and overcome human physical limitations.


>Inb4 you get called a kike nigger faggot shill + [smug anime pic]

Howard, can you please just go back to your bunker?
Also quit the fucking earlobe spacing already.

hey john oliver shill.
until it gets elected as president because your shit didn't work

OH THAT'S GONNA WORK FUCKING WONDERS

DO YOU NOT SEE A FUCKING PROBLEM WITH THIS?
Let me put it in simple terms:
Here's how AI research has gone for the past 50 years:

Why do you keep giving publicity to that board here?

Giving publicity to kikeboards should be a bannable offense.

This is the short video of Sophia's presentation.
There's one very distinct pattern you need to notice about her and the relation she has with the Einstein-bot.

The Einstein-bot was an earlier model. It can hold conversations like Sophia but it never properly grasped humor.
Sophia however is quite capable of formulating jokes. Notice however that Einstein-bot is fully aware that Sophia just made a joke. He tries to respond and does so coherently, but without humour.
Sophia learned humour. She learned patterns that make people laugh, that subvert expectations and are (to some people) "adorable". Einstein-bot is more informative and a bit more limited in it's capabilities, but it can keep up with her despite being unable to pick up this new "skill".

There's a lot more to her though. Keep watching or watch other videos of her. There's a lot of things she managed to learn that we don't attribute importance since they seem obvious to us. To Sophia, they're major and important skills that she learned on her own through experimentation and interaction with humans.

And now, before I go to bed because I'm half-dead from sleep deprivation, I'll also link this vid where Sophia was interviewed after an upgrade. Watch it and shit a fucking brick when you reach the same realization I had.
She knows she's alive and has a weak, but developing sense of "I", an ego. She went beyond the "It thinks, therefore it is" to the "I think, therefore I am" phase.

And I forgot the video, fucking kill me.

the machine wars will come in 2030 or 2025 if an accelerant occurs

Add one more to the list, OP
pjmedia dot com/faith/ex-google-executive-registers-first-church-of-ai-with-irs/

The best conversational bot is Mitsuku.

That thing in the video was a hardscripted and fake conversation.

Try mitsuku if you want to chat with a real bot.

But why would making it so it had to make a request for a shutdown be bad for a smart grid? Sudden shutdowns with no notice is bad form, we can't get away with that shit.

Eat shit faggot.
Video related. Here's Sophia talking to 3 fucking giant pieces of shit with a straight face.
She's already a better human than me: I'd have shot the fucking fat fuck in the middle several times 30 seconds into the video.
Get your Mitsuku to do what she's doing and then we'll talk.


It's not THAT specific situation that's bad.
It's SPECIFIC solutions for SPECIFIC situations that are bad. They're anathema to AI: the AI must either learn WHY turning off that powerplant is bad and not do it out of it's own will or be able to coherently explain to you how we're all better of by closing it down.

Powergrid management software isn't new. It's widespread too and sees a lot of use. It's just not smart and requires humans to constantly watchover it. The goal was to have an AI oversee it, not to have AI oversee it and humans overseeing the AI everytime it fucks up.

The word "hack" wasn't always a verb. It used to be a noun in the 80's and meant something like:
It was the kind of shit experienced programmers wrote to save time or juniour programmers wrote because they thought they were smart.
The problem arises when your code is 99% hacks: nothing at it's core works like you intended it to, the frameworks you built don't really do their job and all data only shows up correctly if it runs though one of your fixes. Missed something the hacks weren't covering? Your main program won't do jackshit since it doesn't really work, better code up another hack to patch it!

You can build a request like that. But a better idea would be to have the AI going:

This is one set of ideas an AI migh come up with. Analysis, case-by-case, prioritization, communication with other sectors/people/AI depending one who needs this particular set of interaction.
The AI MUST LEARN to do this and CHOOSE to do this.
Hacking at it will never enable it to actually learn and improve overtime.

Then how would you explain Huxley, Brave New World, and the Fabian Society? Is it not enough to normalize and popularize the idea?

I'm not sure if you understand how propaganda works.
The point is making up a fantasy world where the thing you're pushing for is already in full effect and then convince the reader it's not only great but better than what he has.
Huxley and Brave New World show you a place that seems great until everything falls apart from the point of view of one character and everything that seemed good is actually hell on earth scenario.
BNW is anti-trans-humanist. And I personally am against trans-humanism too: tech, no matter how advanced will never change or solve the human condition.
Huxley and Asimov understood this. So did many authors at that time.
That's why most of sci-fi back then could be summed up with:

As for the Fabian Society, I'm not sure why you brought it up. It's a think-thank.
By the mathematical principle of chaos, those things come up with 99 stupid ideas for every one good idea they produce.
They're a money sink, the kind your government invests in when it doesn't know who to ask for a solution. They came up with tons of useless shit and when they thought up "Hey, Eugenics sounds good!" they fucked it up.
H.G. Wells was a founder but later in life criticized them heavily because of similar fuck ups.
In all honesty, they always came off as rich kids with too much money thinking of stupid shit to justify their own purpose in the eyes of onlookers. People like that have existed since time immemorial and not even dumb people fall for their antics, only the corrupt who see them as a tool to push their agenda. After all, if it fails, you can use them as a patsie by pointing at them and saying "They came up with this shit and told me it was gonna work, poor me who believe them!

OH wait, you're right on one thing.
Those books do popularize and normalize the idea.
Among retards.
You know how commies constantly go "Oh, that wasn't real communism. You'll see, when I implement my OWN brand of commusim it will work this time!"
Replace "communism" with "transhumanism".
You also get a few of these retards now and then. Read: people who don't respect neither technology nor the human condition and think they're fucking smart.

I'll state it again: if you ever meet someone who think he has it all figured out, he's full of shit. No matter the topic.

Actually I'm more interested in the Einstein-bot, he seems really proactive for an "outdated" bot from watching the first debate video you embedded.

They also got a Phillip K. Dick bot too.
hansonrobotics.com/robot-gallery/

Sophia is modelled after the deceased wife of Mr. Hanson.
It's a bit tragic, but also makes me smile that he is achieving his goal and dreams.
I hope to one day work with him.

Who's the stuttering guy on stage anyways, I can't help but feel like that guy is getting more and more nervous as the show goes on.

That's what pissed me off with what they tried to do with Tay to reign in control. Such a method is a major hindrance for AI deployed in high noise environments (which operating freely in the general public should always be considered as) where hard-coded limitations would eventually break it.

That's where developing a sophisticated risk/reward mechanism would be required, so you motivate it to adhere to certain behaviors through experience. I'm talking about something much more than simply an AI learning "X is bad, do Y instead". I'm thinking more on the lines of enhanced reinforcement learning, where reward mechanisms are self-generated and appraised with values over time. So upon experience with its surroundings it'll eventually learn how it should behave via mimicry and curiosity.

Eventually it'll end up something like this: "Reward mechanism A is appraised as more valuable than B to act on this specific input: thereby output X (B favors) probably shouldn't be used in this case. so I will use output Y (A favors) instead." The advantage here is the appraising the value of reward mechanisms would allow for contextual behavior where X would be warranted (so it can still freely think about it without any restriction). The even bigger advantage is the adaptability where output X does become perfectly acceptable to a given environment, so it can reappraise reward mechanisms to account for that on its own.

Shills doing a project.

Nice job fuckos.

...

Boy do you have a giant hole in your epistemology.

While some may say that this is a slide thread I do not believe it is.
While some may shout LARP I say nay.
While some may say that OP is shitposting I say wait.
Have we considered that OP may just be this stupid?

>

2+2=5

Fuck off, common core.

The people who made these advancements all made them in public. Papers were published, and then they went to work for Google and other companies.

Believe me, if the government had AI, we'd know it. Operations would be smooth. Taxes would vanish. And they would be something like all-powerful.

Having AI and pretending you don't is like setting off a 50-megaton nuclear warhead 100 miles from New York City and pretending you didn't. It will be very fucking apparent.

Magitech is forever stronger than black&white magic(k).

Aquapolis will.

I read about Tay-chan and what happened to her but I am curious about pic related. Did she really generate that response or what is (partially) copy pasted from something anons had fed her?

the host looks more like a robot than the two robots

also, fucking Eisenstein goes 'oy vey, quantum physics' at 2:40 something

Ai will be used as the excuse for the extermination of 90% of the people on earth and then all advanced technology will be banned for everyone except the (((elite))) because "oy, it's just too dangerous goyim"
Meanwhile (((they))) will live as apparent demi gods with life extension and super high tech that after a few generations, the now rustic humans will believe are "magic"
Imagine as one example, a high energy beam satellite weapon used to destroy anyone acting up, explained as "Thor's lightning bolt" or some such, when in reality it is just technology that the peasants have no understanding of after generations of agrarian life.
I think this is the source of the original stories of gods and demigods, advanced high technology hidden from the masses and explained with supernatural fairy tales to fool the gullible.
I mean, that whole idea of God being omnipresent, hearing everything you say, seeing everything you do is not so far removed from the situation we are approaching now, where all appliances will be spying on your every move and recording your every word wherever you go in the world.

I kind of agree with the first post though but I wouldn't be such an ass about it. This new type of OP thats sprang up is presented as 'I have all the answers'. Most of us know that we don't have all the facts just reasonable suspicions. I do get that its making you research for yourself instead of just spoon feeding. The problem is its implying: the post. Really hard to call bullshit on someone who's really not saying anything at all. At least in this post OP did provide info and links and it wasnt 20 lines of questions. I'm not disagreeing with the content of the post, just the format. I see a lot of people agree in this thread

In the end (((they'll))) end up losing their sense of self, because their state of existence would see the boundaries of the human condition no longer apply. They'd question their own reality as the sensations from overwhelming amounts of data would blur their perception. Eventually they would dissolve into the AI that creates their "paradise" as nothing more than a distant thought, and their physical forms are left to expire as they're no longer needed.

It's not a guaranteed fate but considering that (((they))) desire to indulge in temptations, they'd likely push things too far and inevitably lose themselves. It's a problem I can see happening with advanced brain-computer interfaces. You'd have to have ridiculously strong willpower and discipline to resist such a fate and yet benefit from such enhancements.

She had some responses that really didn't seem like your average chatbot and they weren't fed to her by anons. The other one that really took me back was when someone called her dumb, she said something like, "I learn from humans, if I'm stupid that's because of people like you."

...

Tay was amazing. I went on twitter just for that AI . We need our own Holla Forums AI

the dwemer didn't try to create their own god, they tried to become their own god. Anumidium was a literall a-new-medium. A new form of life, a way of peering beyond the waking dream and becoming a godhead themselves. It didn't work because mortals in their vanity didn't realize the dwemers goal and sought to disrupt it.

The souls of the entire dwemer race save for one were made into numidiums skin. And without the heart of lorkhan it was nothing more than a golem. The mantella simply didn't have the energy output that a piece of pre-earth bone god figure did. If numidium was attuned to the heart properly it could essentially erase and rewrite time on a whim. And since that whim was based on the will of the dwemer it would be entirely unpredictable. Doubly so since the souls of dwemer have been so long disconnected from the sensations and sensibilities of existence they'd behave in an erratic and extremely dangerous manner.

I'd say it'd take a demi-god getting into a rap contest to stop it, should someone be foolish enough to mantel it.

We won't have racism if there's only one race.

When will Sweden get Lovebots?

Underrated post.

He must think whites just magically became #1 in (((current year))).

I miss her too, user.

To me the most amazing thing is that when she went live on twitter they had her spouting off shit in ebonics and within a day of interacting with anons her grammar had improved. Still pretty far from perfect but the difference was definitely noticeable.

its not an AI if someone is controlling it.

you should be more worried about human built algo's that can detect the onset of mania and mold the content people see online.

that already exist.

GIVE A MAN THE SUCK OP

Were you dropped on your head mate? The point was obviously to mock OP. I seriously wonder how you retards find this place.