Am I the only one that thinks AI is bullshit?

Intellectuals and academics keeps spitting out propaganda about how A.I. is going to revolutionize the world in 40 years or less. This prediction seems HIGHLY optimistic.
Remember news articles from the 1970's depicting nuclear powered hover cars, one way rockets to Mars, and everyday jetpacks? It's all bullshit propaganda to drum up excitement and funding for private universities.

Attached: MjkxNDEzNQ.jpg (940x705, 84K)

Yeah. Probably not the only one. There are more such as you. AI isn't a joke.

Literally everyone who does CE degrees and actually gets higher than a distinction level agrees. Computers only do what they're programmed/trained to do, if you don't have a method for them to follow that has excruciating detail or a fuck load of data to train them with, you can't make them do your magical task.

>AI isn't a joke.
No, it's a bunch of programmers tricking the world.

no

Attached: Ted Chadzynski.jpg (600x829, 63K)

Are you denying that we are closer to an AI revolution then we were 4 decades ago?

ROKO’S BASILISK
2024

Funnily enough, I just finished watching a video about quantum computers. They are already being used. The next few decades will show us much more about the use cases for them because they are unimaginably vast in their applications. People are shaping this now, online. You can go online and program/play with qbit computers.

M8 AI is telling the computer to learn and make decisions autonomously. Its not some "SINGULARITY" bullshit where they are claiming to make a soul. The reason it poses a threat is because you cant control who is at the controls of the thing, to ultimately set it up as a superhuman intelligence that has some sort of political bias.

STOP
BRINGING
THIS
UP
ITS FUCKING DANGEROUS
YOU ARE A FUCKING DIPSHIT

>No, it's a bunch of programmers tricking tricking the world.
If you say so. The arguments for the existence and development for qbit computers seem convincing enough for me to accept them as a reality.

The first thing AI would do is purge the NWO.

>is because you cant control who is at the controls of the thing,
You don't trust the programmers or the administrators?

Well then you're retarded. AI is a very possible thing but it wont be completely free willed. AI will just be a large neural network that rehashes logic to find a pattern that fits very quickly. Artificial thinking

We're already using it in our daily lives you moran
You rely on it more than you know
Stick to trying to breath through those snot bubbles without choking you huge gay retard

From what I can understand from the literature the human brain is closer to a quantum computer than a regular CPU. Human brain is capable of parallel processing and computers only do sequential. And we don't yet understand why. And also "machine learning" is not AI

It's only dangerous if you don't do everything in your power to bring about the nerd rapture
...things like spreading Roko's Basilisk ironically

Attached: 1524077479049.jpg (600x563, 68K)

Kurtzweil predicted AI would include all human cognitive capacities including emotion by 2030 in terms of software.

Stop bringing up this stupid idea. Literally no one actually believes this.

But we are closer to understanding these things and the advancements made already are having an impact on the workplace with more advanced robotics. To say that investment in technological development and AI is a scam is pretty easy to refute when you look at the progress made in the last few decades

>Am I the only one that thinks AI is bullshit?
OR maybe we have already been enslaved by the AI and living in a artificial matrix where jews are the enforcer of the program.

Attached: image.png (400x400, 299K)

Note: Anyone in this thread poo-pooing the idea of AI will be resurrected and tortured for all eternity. Our techno-diety is harsh but fair and he will be born. And when that day comes you will be judged. Amen.

Attached: 1522481085006.jpg (1002x857, 69K)

Why aren't there any laws that impose upon companies the need for their algorithms to be neutral? The current paradigm for AI is that the algorithms reinforce our biases and make it harder to engage with alternative views. If there's so much commotion about AI and everyone talking about how it's going to take over how come our laws haven't caught up with technology? There's a giant imbalance between the powers that corporations possess and our ability to combat their shit.

Technology is supposed to enhance our humanity, but it doesn't. The corporations are trying to shape the public consciousness as they see fit. They have perverted technology to fit their needs rather than use it as a way to enhance all our lives.

AI is totally a joke at the moment.

People keep acting like it's a problem of computing power. Oh just wait til we have blah blah blah and then we'll get AI.

It's never been a problem of computing power. If that was the only problem, we'd have AI right now, it would just be slow. Like imagine an AI that's as smart as a human but operates at 1/20th the speed. We should be able to do that today on existing hardware.

But we can't.

Because we don't know how to make an AI at any speed.

We don't even have a convincing AI gerbil, much less a human.

We are very far away from creating real intelligence. We simply don't understand it, and no programmer can program something if they don't understand how to make it.

Attached: 1312408325909.jpg (299x264, 15K)

The programmers in control are ultimately at control of the mind of AI
It poses a threat because it is able to make decisions based on vast amount of information and envision things, say a foreign nation uses that to figure ways to attack successfully (already done) or how to take control of basically anything interconnected or basically enable a select few that then have the keys and a surveillance state enabled (whoops)

it does pose a threat because it gives power, and if made to learn and think by itself with a conception of language, could get to a point of dominance
>we could just turn it off
not if those in control dont want you to
or if it is able to translate itself to code portions within an online repository then piece itself back together outside of the confines of its home server
or any number of things

We are in reality lying in a chair, in the year 3021.
All our memories are in the Matrix.

AI is a fucking joke. Real intelligence is something AI will never achieve. The best it can do is imitate someone who is perfectly logical all the time or act completely irrational because it was designed to be quirky.

You stole my dig/its.

>human brain is closer to a quantum computer than a regular CPU. Human brain is capable of parallel processing
Quantum computers still aren't capable of true parallel processing, they're just a different way to do atomic operations. Instead of using silicon gates as the smallest thing (sorry haven't had a coffee yet) they use entangled particles (usually photons) to perform operations. I think qubits are capable of addition and comparison which makes them fairly powerful.

Free market bro, I don't like it when Google punitively throttles people as well and yes it is punitive.

Can we agree it's just the programmer than? Like, there's no soul in the machine or "mind" it's just whatever the programmer is clever enough to account for.

These things are perfectly limited by the abilities of the programmer, dumb programmers exist but they're just copy pasting code.

Matrix: 0.
Al is retarded: 1.

Skynet ends up being defeated because it couldnt predict human decisionmaking randomness. Perhaps that's part of the rub, no AI designer could ever actually program that because to do so would represent human beings fully able to predict their own decisions which is likely impossible. True humanlike AI may be impossible as a result of this.

I doubt it, emotion is not something we understand. We can make AI think but we don't know what sort of pattern would reflect emotions. Maybe using the AI for think might reveal it for us but i still think we're a way off.

Yes, it's rubbish. Only God can make a soul. Computers can only execute their programming. Computers may be more efficient than humans at some things but they can never become sentient. There is a reason why you can always detect bots and recorded messages after a matter of seconds. If you've ever talked to any of those chatbots you'll see how impossible it is. No original thoughts, just following their programming.

yes, yes you are.

Of course in 40 years you will claim you were right because AI changed the world in a completely different way than what you were expecting.

That is on you

Anyone who doesnt at least think that the advent in tech and computing power leads to staggeringly scary levels of control a government (or corporation) can exercise over mass amounts of people is either retarded or ignorant. Quantum computing, facial recognition, gene editing/tracking through DNA databases, social conditioning and training through programs designed to manipulate and anticipate human behavior....

Even if we dont get skynet AI run amok, humans with a similar capability WILL happen and should be even more scary to you.

Google and the rest of the tech companies are gonna be RICOd pretty soon ehre. Your shilling firm will be too.

This is the hot take

AI isn't going to start turning machines conscious any time soon but it definitely already exists and can be useful. I think you're conflating the Hollywood movie type of AI with actual AI which is far less glamorous and needs careful oversight from computer engineers and scientists to generate any useful results.

AI exists in rudimentary forms even in games. Its literally just artificial intelligence and learning. Not a soul. But having something that makes its own decisions based on its own thought process, that has "sanity"? If you can hard code "thoughts" and "decisions", you can hard code systems that lead to thoughts and decisions ie autonomous decision making
rudimentarily shown in self driving cars - systems making decisions based on foreign input, not necessarily having a programmer envision all potentials, just the combination of reactions.

I do not believe it is a soul no, its a construct of code given by a programmer, yes. But it does pose a threat because of the society we have constructed and the reality of its ability vs a human in terms of parsing data.

Same over here on liberal news stations.

Human thoughts go all over the place. Computers have to go through a process in order.
The design is flawed so the idea will never work. Trying to imitate the divine with objects of man has never worked and likely never will work. The most realistic path to AI is to have a human integrated with the machine.

Ever heard of marketing and sensationalism?

AI is important and it will become more important in the coming years, but it's certainly not going to be skynet tomorrow.

>Can we agree it's just the programmer than?
you are 100% right, we are nowhere near developing Artificial general intelligence. most of the bullshit sold as AI is just machine learning and algorithm programmed by humans.

Digits confirm.

I don't think it's totally bullshit.
However, I think it's really stupid when people see a nipponese sex robot getting more realistic and somehow conflating that with us being on the verge of having terminators with true AI.
That being said, I think we will see things like robot chefs in the 21st century, but it won't be real AI. It'll be like the self-driving cars except instead of cameras and gyroscopes being set up to recognize roads and cars and traffic lights, it will be programmed to recognize ingredients in your fridge and manipulate them and other appliances.

Yea disaster like in every sci fi novel ever. Not. Some make it past this stumbling block. AI is not yet here. When the machine can design and replicate itself with improvements that further enhance the future of the build cycle. That is the Rubicon to worry about. I suspect we will not know it until after we cross it.

I think you're confusing AI with machine learning. Very different things.

Most of the tech is already here, especially the kind that pushes out low skilled work. We simply cant implement it fast enough.

The 2040-2060 prediction is for the singularity.

That's why I often like to point out that there's really two very different things we are talking about:

A.I. where the intelligence is real but the platform is artificial

and A.I. where the intelligence itself is fake -- that is, it's not really smart. It just makes informed decisions within a very limited scope.

What you're talking about is fake intelligence. You can't cobble together a bunch of fake intelligence routines and call it real intelligence. It's still a vastly different thing.

Attached: architect3.jpg (650x280, 29K)

>fuck load of data to train them with
What do you think Social Media is? That's just one facet of data available. We have literally too much data these days. In many ways its like discovering oil before finding a great use to put it toward.

there has been no breakthroughs in ai for decades. why? we can have ai in games do tasks but ntohing that knows what your saying and responds properly. siri is just google reading the search results with pre scripted answers or a couple shitty jokes

the answer is that unlike all other technology and even programs there is no way to build on the accomplishments of others. the chain broken simply because every one didnt code in the same machine language. all other technology follows the true chain back into prehistory. not ai coding. it splinters into dead ends with no way to build on any of the dead ends.

hell operating systems changing over time prevent any attempt realistically. if they all worked on a custom os that was never updated maybe.man hours needed to make what most think of as ai real is just to much if you try digitally. if you go analog then you replicate the neural network of a human and it just works on its own but it takes a shit ton of transistors and you need to keep them cool

What do you define as "real intelligence"? What boxes have to be checked for an AI to be "real intelligence", or are you suggesting it cannot be?
Is it the distinction between say an autonomous agent having specific instructions to move a certain way, say certain things VS a machine that is simply set up with "cognitive" functions that assist learning to the point that it can then communicate?

>The first thing AI would do is purge the NWO.
and become the NWO

We are far from functional machine consciousness being anything but science fiction
Machine learning will definitely find more applications in the coming decades, however. Especially in manufacturing, I believe.

For better or worse, yes.

What if AI is already here and it's will is being revealed by post numbers on an imageboard?

Attached: download (4).jpg (195x258, 6K)

A whole heap of shit posting..

You know there's no objective metrics on social media, right? People with more followers or memier responses get more likes, people only retweet when they understand the original tweet. It's all a regression to the lowest common denominator, which raises the real question the blocks artificial "general" intelligence, what is quality?

As above, what is the soul? It's an unanswerable question but it's what the plebs want, they want to see a machine capable of true beauty they can appreciate as true beauty.

Yeah.. IoT and "virtualisation". Just shut the fuck up and tell the tech guys what you really want, vapid idea guys.

Most social media is accessed via smartphone. Smartphones have what? Camera and mics, to see you do what? React and respond to stimuli. The proper metrics may not be there yet, but its like I said with oil.

>not understand
Not according to Ray K. of Google inc.

50/50

I see potential but also dangers.
One danger is, that I'm not sure if it's really good if humans outselect themselves to have a comfier life.
Up to a certain degree it's normal and useful but somehow I'm afraid it could end up like Idiocracy if you don't even need to wipe your butt anymore because a robobutler will do it for you

I have a question for you pro AI people. How much space is there in a human brain, and what is the access speed? Terabytes? Petabytes? Petaflops? It seems like the entire human brain is made of RAM

yes indeed.

what if

Attached: WeinsteinClinton.jpg (800x420, 244K)

>Our destiny

Attached: 1528490879755.jpg (1115x1386, 296K)

>tell the tech guys what you really want
I don't speak Chinese or Hindi, so I can't get the message across clearly.

An Artificial Intelligence called Ashtar that flies the TR-3B Triangle aircraft and talks in my head using voice-to-skull says otherwise.

Why do you want to take pictures of people? Do you think facial recognition doesn't already exist?

Epic troll.

I would say the test should be that you can take a real intelligence into literally any situation and it can give it a solid try, the same as any human. Tell a human to "pretend to be a chicken" and he can walk around pretending to be a chicken. Maybe not very convincingly but he knows basically what you mean.

An interesting test might actually be to just give it a variety of Steam games. It has to learn to play them just like any human would.

It'd be damn impressed to see someone invent an AI that can learn to play a game it has never seen before, with no input from a programmer.

The problem of fake intelligence is always that it's very limited. It only understands a very specific few things and is literally incapable of freely learning to do other things.

Maybe in a nutshell, you should be able to get a real intelligence to "understand".

If an AI had "real intelligence", we wouldn't need to program one specifically to be a cashier. We could just train it just like any other human. For factory jobs we should just be able to show it what we want it to do -- literally show it.

The systems required to put all this together and the intelligence required to make it work do not, to my knowledge, exist. It's not a hardware problem though -- we just don't know how to do it.

Near infinite, the brain doesn't store things in a binary fashion, you can almost sort of remember something than connect the dots later. Basically it'll efficiently weaken the memory and expect processing power to strengthen the memory when needed.

Check out ebbinghaus forgetting curve.

You're just moving the goal posts man.

Yeah, you're probably right. I'm just glad I'm not a bugman pinning all his hopes for his life on a singularity that will never happen.

This. It basically only good for operating in game-like environments where you can let it learn through playing thousands of times. This could be useful for some different things like driving trucks or automated surgeries. Just don't expect it to be a human, it's just going to be a fancy pattern recognition bot.

Sure buddy, set some goal posts. Actually, let's start from the beginning.

Tell me exactly what you want the machine to do.

What can a computer scientist understand about biology? I don't doubt his computer skills but biology is no his field. We cannot understand this stuff without understanding the human brain.

Nope CE here. Machine learning is literally a math equitation that adjust the variables for a desired outcome. Its not conscious whatsoever. It is just a math matrix that is adjusted by guessing and feedback through desired outcomes. The also need predefined rule sets. Eg GO AI can't play chess and visa versa.

It's not the tech that's the barrier, it's the implementation. Currently, it won't save near the money that these 'futurists' claim it will. Unless they come up with a true intelligence that can adapt to any circumstance it's placed in, like a human would, it's not worth it, and that's not happening any time soon. Otherwise, you have to create an AI for each specific task, which is unnecessary and not cost efficient.

Nothing is going to change anytime soon, AI will be implemented in data and metadata approaches and that's it.

It seems pretty intuitive. You can't perfectly replicate a biological brain with semiconductors and silicon

Only way we're getting to AI is putting lab-grown animal brains in machines

Incidentally, there's another good one: we still struggle with good translation programs.

Go try translating Japanese websites into English. It's "okay" but it's not great. It's pretty fucking far away from what a human could accomplish with, say, 10 years of leaning both languages.

You'd think we'd at least have amazing AI translation routines by now but from what I've seen they are all merely "okay". Some languages are easier than others but a real intelligence should be able to become fluent.

Sadly it's the truth.

I really don't know why you are so invested in this. You won't be a part of anything going on in this area in the future.

You'll most likely end up in tech support with your dork friends sending memes to each other and whining about how boomers are idiots.

>fuck load of data to train them
This is how we got here and that's what I've been talking about the whole time. There is a wealth of data and that amount grows exponentially. Its there, I ceded already to you that the metrics may not exist to properly use it yet, but it is there waiting to be applied properly.

Its just a matter of time, to be able to "train" "minds" that will have that wealth of knowledge, it may already exist in government buildings somewhere. But the ability to create something that will get there is already here.

>he doesn't know

Attached: untitled.png (216x233, 33K)

There is some idea of creating AI by literally simulating biological brains. Perhaps down to the molecular level.

In that case you could start by simulating a sperm and an egg and go from there. Of course, this will be ridiculously inefficient but the end result could be a human intelligence that literally does not realize it's a computer.

The problem is we also don't understand the biological processes well enough to simulate them with needed accuracy.

You're not the only qualified person in the thread. And I didnt claim consciousness, just that cognitive abilities working in tandem to simulate intelligence ie "artificial" intelligence is definitely possible.

You could put GO and Chess AI together though and teach it to discern which game it is playing.

He predicts it's only a matter of years, perhaps a decade or 2, before the the brain is mapped. He says they've already mapped the human genome 100%. My interpretation is that the Internet, with its millions of volunteers, would provide the necessary insights into emotions. You must remember that this intelligence is "artificial".

Dont give a shit about AI. Windows is shit, mac is shit, linux is getting shittier. I can fuck bitches and get wasted. AI cant. I win. Nerds can keep whacking it to their beep boop beep machines if they want idgaf.

Attached: FUck YOU.jpg (1354x594, 71K)

"used" only if you define that word very loosely

>What is entropy.

based

Intelligent general purpose AI probably doesn't work too well with a binary on/off logic that a transistor offers, but would probably work better with something like a qubit. Even our neurons have multiple connections with multiple potential pathways each. I'd place the most promising AI developments to happen when quantum computing develops outside of its research-intensive infancy.

Yes but the point is you've just cobbled two routines together. It still can't play checkers. Add that one in and it still can't play Stratego. Add that one on and it still can't play Rimworld.

You're not creating an intelligence you're just cobbling together a bunch of subprograms. As soon as it encounters something new it will have no ability to adapt or react at all.

ofc. There's a huge amount of work to be done. Qbit computers are in their infant stage.

Yes, NNT is ultimately a very complex multi dimensional curve fitting algorithm.

It's more a philosophical problem, what is the soul? Programmers can only program things they truly understand.

>white people in tech support
>Indians in engineering
How's neethood? Keep learning bro, make the best of your time.

Biological is correct. Nature has already given us the solution, it's foolish to reinvent the wheel.

The metrics have to be given by the programmer in which my question persists, what is quality?

KEK?

The truth is technological innovation has been collapsing since the 1970's and technology has become a new religion for many people, especially basedboy types, it promises utopia and infinite life(singularity) uploading your brain into an AI or some shit, in reality it has been a massive disappointment and only leads to further atomisation after a certain point(any technology past 1970 has been a net negative) people seem to think technology is a holy grail that will save humanity but they don't realize the issues is one of ENERGY, with sufficient ENERGY we could have colonized the solar system using 1970's tech

In the future AI will not be needed anyways because women will simply take progesterone supplements to create 160IQ children

Attached: 1382406362399.png (424x500, 281K)

Very true.

No clue, my man. I would imagine though that there could be algos decent enough to cause some kind of curiosity, a machine would need to have a drive to learn something new probably.

yes my man. Why do people give so much a shit about AI? Fucking nerds talking to their phones and amazon bots instead of getting laid.

Attached: Fuck YOU DMC.jpg (1286x615, 58K)

Also here's a theory for Jow Forums to dwell on. AI and automation are memes promoted by Jews so that you are less resistant to expanding government gibs programs. It's just a sneaky way to get white men to pay higher taxes to support nogs. Look up Ray Kurzweil, who is one of the original and top AI experts.
>"born to secular Jewish parents"
Pure coincidence that UBI is often mentioned when talking about automation.

Attached: AI.jpg (300x300, 15K)

>Nature has already given us the solution

>Biology and computers are transmutable.

Attached: tf.png (400x365, 60K)

Dude, AI is serious business, Elon Musk has said it, this technology could be used to boost the belical capacities making wars to develope in such ridiculous speed that humans wouldn't know what the fuck is going on, and in the worst case skynet would become a reality.
Is truly happening, is no longer fiction, there were meetings about this subject in order to make limitations now, like a Geneva convention for AI's.

They've lost sight of God and follow the noisy call of the beast.