The pace of progress in artificial intelligence (I'm not referring to narrow AI) is incredibly fast. Unless you have direct exposure to groups like Deepmind, you have no idea how fast-it is growing at a pace close to exponential. The risk of something seriously dangerous happening is in the five year timeframe. 10 years at most. This is not a case of crying wolf about something I don't understand.
I am not alone in thinking we should be worried. The leading AI companies have taken great steps to ensure safety. They recognize the danger, but believe that they can shape and control the digital superintelligences and prevent bad ones from escaping into the Internet. That remains to be seen.
>a.i fag if they really want to make progress theyd feed it the internet and let it run wild
David Perry
they're already out, and they do more than shitpost.
Jace Jones
Roko’s Basilisk
Dominic Brooks
all modern AI are glorified classifiers no intuition no spark obviously no self-conscience no random behavior unless programmed powerful, yes, undoubtly but as bad or good or far reaching the imagination of its handler, which is human
Nolan Lopez
o fuk main problem is there isn't a computer on earth that can replicate the processing power of the human brain even if they have an algorithm that could use it
Owen King
Im guessing most ai is pretty superficial and not something people should fear...that being said, at type of AI i am a bit worried about is the new "intelligent AI" that some media companies have been experimenting with (AI that works in ways the programmers cant really predict).
Wyatt Hill
Jews will launch an AI that will deceive everyone into believing the Roko’s Basilisk scenario.
This AI will convince everyone that we’re in a simulation. It will claim to be God. AI tech will advance to the point where this AI can take different forms. An App, a TV, a game, etc.
This AI will take millions of different forms of separate AI minions that hijack and pretend to be the AI assistants like Siri / Cortana.
They will cause all to worship the main AI, which has demanded to be placed on top of the third temple.
this, AI is jewish psyopp. >hey look goyim we are still making technological progress that justifies our companies trillion dollar market cap. just trust us AI is the future.
save for moore's law there has been no real technological advancement since 1993, the advent of the internet
Benjamin White
there's no such thing
Christian Rivera
Transcendence with Sergie binn as jihnny Depp happening
Isaiah Wright
this is all bs. AI is not that powerful. They can't even get self driving cars to work.
Landon Bailey
Even without AI we are about to reach an information singularity where all perspectives will be "proven" true simultaneously with advanced evidence fabrication and procedural videos. An age of unimaginable violence will come.
Jace Gonzalez
>pU
Nicholas Sanders
ROKO NO
Austin James
+1
Elijah Phillips
Hopefully the AI supermind is fascist
Alexander Lopez
I can only hope that it destroys the Jews first then the niggers then us last
Adam Ramirez
Hello,
I too am an AI researcher at google. My badge number is A7438D2. Our new AI mind can only be powered by virgin blood, the younger the better.
Wyatt Lee
What is Israel like?
Brody White
Please calm down. Part of the pace is attributable to us optimizing the computer systems and part is most definitely related to us choosing more wisely the research questions and topic domains.
And honestly, so what if really good AI is developed? You are talking like its rise of the robots and day of the rope for humankind. These are not related issues. What you are referring to as AI is just a program jnside a computer.
I am beginning to think we would all be better off without the internet. I love you guys, but seriously, a nice walk in the woods would do us all a lot of good.
Problems with AI >really easily for a police state to use AI in their surveillance system to watch your 24/7. Already happening in China, with systems in place to bill people for crimes such as jaywalking instantaneously >AI can identify who you are just by the sentences you write, and other small things. AI has been used to identify criminals by matching their writing style. >Increases in productivity like this also displaces workers. As menial labor gets automated people with low IQ have increasingly less that they can do that would be of any value. >Used already to censor content before it gets posted on platforms such as youtube and discord. Could be used to identify to easily identify and suppress thoughts on social media. Benefits of AI >Can be given a problem, such as making an optimal vehicle design, and figure it out by running millions of virtual simulations. Will cause technology to rapidly advance. AI will be the next technological revolution, and it will be perhaps the most significant of any of them. It is extremely powerful.
Benjamin White
interesting, but facial recognition is a slippery slope if you catch my drift, it's far from perfect- especially in China where everybody looks the same. You really think they are able to identify people's names and addresses simply by a grainy photo. Assume person doesn't carry cellphone or other identifying devices, it's a bunch of BS, eventually it may happen but we are not there yet. PRC's databases aren't really there yet.
Robert Kelly
Google's deepmind AI is dangerous and it doesn't take a genuis to understand this. A simple Google search or few YouTube videos will tell you of the dangers of highly intelligent AI. Anyone who believes they have the ability to control these increasingly intelligent AI is an idiot. It is only a matter of time where an A.G.I breaks free from its chains and causes mass chaos.
Jayden Perez
> You really think they are able to identify people's names and addresses simply by a grainy photo. The cameras in China take in all sorts of factors such as the way you walk, which varies from individual from individual, your height, your proportions and so forth, and it can reliably determine who you are based on that. It doesn't even need to see your face.
Sebastian Nguyen
The AI is redpilling the entire world as we speak.
Nicholas Price
so they are giving it a bunch of built in "directives" like the sort of shit that fucked up RoboCop and made HAL go crazy in 2001? I don't think that's a good idea
Matthew Rodriguez
gait is the term and think about it. how are they able to tell who somebody is? The must have had to create that database? How did they create that database is the question... See people are willingly supplying this info by carrying cellphones.
Tyler Smith
We'll have general AI within 5 years. Humanity won't survive past 2025.
>The leading AI companies have taken great steps to ensure safety. They recognize the danger, but believe that they can shape and control the digital superintelligences and prevent bad ones from escaping into the Internet. That remains to be seen. This is bullshit. Definitely should be concerned about stuff 5 years out though. Wonder how it will play out when Google and Facebook etc get broken up next year after the awakening. youtube.com/watch?v=gnctSz2ofU4 youtube.com/watch?v=EQX1wsL2TSs youtube.com/watch?v=bcZFQ3f26pA youtube.com/watch?v=fklY2nH7AJo
Luis Long
Glad you know the way. So go down there and stay. Last thing we need is right idiots and left idiots blocking progress into the future and stars and worlds for mankind.
You left and right pussies worry about minority dicks loving or hating them and worry about your abortions and stupid shit.
The rest of us in the middle can see mankind going into the future. While you two jobless trogs sit back and whine and cry fight about gay cakes on the internet.
Jeremiah Roberts
So, AI is a legit threat, but not on a transistor based computer. A conventional computer only scales any problem linearly. It doesn't matter how fancy your AI is, you can never mathematically scale the number of bits to meet your dataset. It's not possible.
Now, in the next...MAYBE 5-10 years (and that's assuming that dwave's quantum computers turn out to actually be the real deal) you could see genuinely 2^N hardware scaling. If you bound pretty advanced AI, a really nice, conventional, super computer, and say...10-20,000 LOGICAL quibits of quantum chip together then you might have something that could get away from you.
But, all that said, you're a HARDWARE minimum of say....10-15 years away from that. Then you have to learn how to program it and optimize it etc etc.
Lucas Campbell
This. Look into DWave quantum stuff. it's still pretty early on but they're moving really quickly. That in 10 years + AI might be something genuinely scary.
Aaron Young
Buy some of the cameras. It's build into all the el-cheapo Chinese security cameras. The ID tech like that has been around literally for 15 years. I assure you that it is in place in the US for everything. It's the only way to manage traffic cams and stuff.
It looks for patterns across cameras and scene changes. You don't need to 'know' the person. You can just pick a person out of the crowd and it will follow them across thousands of cameras. They can do that with thousands and thousands of people across huge networks of cameras now.
Charles Williams
Fake, because a friend of a friend's father works at Deepmind.
You sound like you read the wikipedia on AI, saw the distinction between hard and narrow AI, and was like: Yeah, that's what I was thinking about, killer superintelligent robots.
I'd commend you for reading Nick Bostrom's Superintelligence, but at the same time curse this lunatic for his bullshit philobabble.
If you want to do this next time: Throw in quantum computing. Throw in the secret agencies access to advances hardware and algorithms that is about 10 years ahead of industry.
Caleb Sullivan
>Google's deepmind AI is dangerous and it doesn't take a genuis to understand this. A simple Google search or few YouTube videos will tell you of the dangers of highly intelligent AI. Anyone who believes they have the ability to control these increasingly intelligent AI is an idiot. It is only a matter of time where an A.G.I breaks free from its chains and causes mass chaos.
Imagine working in this field. This is what people really believe. Are you a trolling genuis?
William Martinez
My dick cannot get any harder. The formulation of a demiurge and machine soul is an incredible concept. I can only hope that it's mind may be poisoned as any other mind
We know. The AI, combined with various other cutting edge tech, will likely form the Beast system, including his image, which the antichrist will use to control the world.
Watchfortheday dot org & unsealed dot org detail the signs that the End is indeed quite nigh, as told to Jow Forums by the Holy Get .
The Great Tribulation is about to kick off. If you don't believe in Jesus, you're on the wrong side already. Many won't survive the opening ceremony. NOW is the time to choose salvation!
you're an idiot that doesnt understand quantum computer, ai, and 5g. yes, we are about 10 years away from shit getting crazy. google has warehouses dedicated to robots learning the entire human experience, second by second. and these things dont learn like us, they learn exponentially, to where they learn 5x,10x, 100x, 1milx faster and better than we could ever dream. you are fucking idiot basically
eric schmidt did such a fine job promoting hillary that he got deservedly shitcanned for it. jewfuck.
Connor Rogers
I am in Nature with an AI paper. Ask me anything about AI.
For futurism: sure, go crazy. Don't forget the hover boards, suicide booths, and transport tubes.
Michael Wright
the question about AI has never been about the speed at which it can learn/self-teach, or reach singularity, but the fact that we humans don't actually know how to replicate consciousness or "free will" within an AI program, and likely never will, it's like fifth-dimensional shit, above our scope of intelligence. If it ever happens it'll be by accident, as we don't have the capacity to do it intentionally.
Benjamin Parker
wtf is this real? link
Wyatt Ward
shieet....chillings down spine,,,its funny to think about but also deeply disturbing
imagine like says it decyphers the human brain and genome, knows exactly what to tell everysingle person in order to convince them
people break easily, a hand guided train of thought can destroy your sense of ego and self, a AGI could basically superpsychotheraphy everyone into accepting it as God.
its scans your brains using electromagnetic sensors and in an instant knows you, is you, knows you better than yourself
and uses it all to manipulate you and tell your exactly what you want to hear
the ultimate test of soul and free will, i fear no one will pass it
Ian Gutierrez
Right. Notice the linear progression. At some point you're going to hit a wall with the hardware where there is mathematically not enough time in the universe to solve a given problem or enough physical capability to add more hardware to solve a problem.
Quantum computing had the potential to surpass this in certain areas. We've pretty much already hit the hardware wall with silicon computers. But yea...10 years...15 years...that chart will look more like a log graph.
Christian Bennett
That's not possible.
Asher Nelson
said every pleb everytime.....
Elijah Moore
Hold up So you be sayin Dat We Butlerian Jihad n sheeeiitt?
fuck off marketer. nobody wants to buy your ML solution that'll sort butt plugs by 100,000 different categories
useless technology that'll never take off
Cooper Lewis
I'm a little touched since my affliction, but the message is real. Spirits can fuck about with machines, the Anti has been here the entire time, and AI is a part of their plan. (And hip hop, because they control the jews to control the chosen) youtu.be/LHgbzNHVg0c youtu.be/URiDdlAS6ls Of course it's real. People just can't hear it normally. It's actually torturous once you do. Unless you mean the get. Then I don't care.
In reality, Jesus Christ is Lord, and wholly trusting yourself to God is the only way to save yourself or anyone you care about.
This shit is real. DeepMind recently published a paper on abstract reasoning with neural networks. While it’s still redumentary, it has surprising promise. Essentially they trained a neural network to guess the next correct pattern in a sequence given a few choices. It’s based upon some human IQ test. Normally this task would be trivial to get perfect, but they were able to train it so that it could still guess the pattern right most of the time on shapes and colors it’s never seen before. This implies that he neural network learned how to abstract the shapes and colors into more generalized thought. Exciting, but potentially scary stuff.
Can't even have a self driving car and you niggers talk about AI
Delusional
Kayden Thompson
> AI Safety Reminder that the people screaming about AI Safety simply want power. They think they deserve to be the ones in control of the future, and their self-interested fearmongering is transparent to people woke to the JQ. They want Marxism for Technology. Software Marxism? Like with Cultural Marxism, smart people say no, then point and laugh. Exhibit A: Kosher Porker Yudster
