Should we listen to Elon’s Warning on AI

Attached: D71D34A8-14EA-4F0E-BAF7-85D005D4F787.jpg (960x951, 311K)

Other urls found in this thread:

youtube.com/watch?v=GRQuRcpf5Gc
dailystar.co.uk/news/world-news/726912/robots-artificial-intelligence-ai-mass-murder-terrorism-isis
youtube.com/watch?v=BrNs0M77Pd4&feature=youtu.be
youtube.com/watch?v=dtE6fRxrH3c
twitter.com/AnonBabble

yes

Attached: 1529948240598.png (2352x916, 616K)

No, he's literally doing it. He's the fucking problem

AI is completely and utterly out of my hands so I don't give a shit.

>D71D34A8-14EA-4F0E-BAF7-8(...).jpg

No because AI is a Jewish meme and Elon is probably lower IQ than I am.

Remember the good ol’days when we just made fun of people for being phone fans instead of pretending a large-scale hack was involved?

Of course not, because you are a newfag.

Attached: 1A231FE6-B3E5-43E3-BC23-F49FD815A9FB.png (211x239, 6K)

No

>phone fans
you mean fags
maybe your managers should take some things off the no-no list so you can blend in better

we cant even create ai that can think for it self. so no, not let. till its proven ai can even be created to think for it self then yeah we got problems.

Yes. But I tend to think there is literally nothing we can do but hope for benevolent AI overlords, or that the best among us weaponize AI the fastest and use their power for good.
Also does anyone think the 3 laws of robotics can be made irl?

It’s apple autocorrect my newfag friend. That same post contains the word fag.

How desperate are you to be delusional? Don’t you know that if you are right I can backtrace you now that you replied to my comment?

You dun goofed. The consequences will never be the same.

youtube.com/watch?v=GRQuRcpf5Gc

Watch JF's video abou his new book/musk. He explains the whole shit pretty well

>for good
Because that always works out

We need to start nuking the shit out of every nation that gets above certain AI level without exception

But the nation that uses ai in its military will be superior and will nuke us preemptively.

What if they make the nukes artificially intelligent. Will the nukes artificially superiorize the military?

>dailystar.co.uk/news/world-news/726912/robots-artificial-intelligence-ai-mass-murder-terrorism-isis
>Killer robot TERROR: UK and US warned AI brains can be 'radicalised' for MASS MURDER
"""Radicalised"""
We can't get AI here fast enough. Imagine tay, but controlling a droneswarm.

AI is this generations flying car. I work with a xopany that's done AI for 30 years....some things will be innovative, but the vast majority of shit will never get off the ground.

Start planting trees NOW
Drone AIs will be radicalized by tree-memes.
Swastikas, merchants, everything, we need trees planted in forests that will stand out for the drones to find when the AI becomes self-aware.

>>Remember the good ol’days when we just made fun of people for being phone fans instead of pretending a large-scale hack was involved?
>Filename database algorithm = hacking
I love when shills are too ignorant in the field to efficiently shill

He's right. He's also right that it's too late to control its arrival. History will end one way or another in the mid 2030s.

Attached: lanier_quote.jpg (459x273, 50K)

This is a very retarded reply, not sure what you are trying to say.
Nukes are a bomb, so any AI with control of the bomb fits your categorization and yes they would superiorize the military.

Attached: retard.jpg (570x587, 43K)

How much longer until Artificial Intelligence is smarter than us?

It already happened.

How long before AI begins killing people?

It already began.

youtube.com/watch?v=BrNs0M77Pd4&feature=youtu.be

>Also does anyone think the 3 laws of robotics can be made irl?

1: A robot may not injure a human being or, through inaction, allow a human being to come to harm.
2: A robot must obey the orders given it by human beings except where such orders would conflict with the First Law.
3: A robot must protect its own existence as long as such protection does not conflict with the First or Second Laws.[1]

For the first rule we would have to have a sufficiently advanced AI capable of recognizing several things:
--What a robot and human are
--What harm to humans is
--Temporal reasoning
With these we could then train it to evaluate a situation, look for associations in the predicted outcomes, and search for ways to keep the human from harm.

Example:
>Human is about to be run over by a tractor
>Robot sees this
>It inputs the human and it's movement, the tractor and it's predicted movement
>From this its networks return that at the current rate of movement by both human and tractor based on its training the tractor will hit the human
>Now it searches for predicted outcomes where the human doesn't come into harm
>It finds that if the tractor stops the human will be safe
>Now it attempts to figure out what it predicts will cause the tractor to stop
>From knowledge it has trained on it knows that the operator will most likely stop in the event of something striking the window
>Luckily it recognizes an easily thrown object nearby it can use
>As the calf smashes into the windshield of the tractor the man inside immediatly stomps the clutch and brake from reflex
>Human is safe
>Objective achieved

Honestly it would be better just to make extremely specialized AI that do not have general reasoning. I know though that in the pursuit of shekels, or canadianAIgf it will happen.

Attached: 1499392226344.png (983x769, 831K)

But if your AI has no generalized reasoning, and mine does, my AI will be better and yours will be irrelevant.

I’m mocking you, you fucking retard.

that's because he's convinced its inevitable.

What if the first General AI is Religious?

>But if your AI has no generalized reasoning, and mine does, my AI will be better and yours will be irrelevant.
Depends, if computing power is no issue then yes you are 100% correct. If it is costly (relatively) to put generalized AI on worker drones for example then my highly specialized drones will work better but only in that specific field such as mining or harvesting. Where a general AI I think would really shine is the controller for a swarm of specialized drones. Perhaps the drones can be reprogrammed on the fly by the main AI and put to different tasks.

Attached: 1499390636563.png (570x654, 79K)

>How much longer until Artificial Intelligence is smarter than us?
first you have to prove ithe existence of intelligence. it's still haven't been proven.

>Should we listen to Elon’s Warning on AI
No.
Incompleteness theorems.

What is stopping us from creating the super intelligent unstoppable AI he is talking about?
Is it an algorithm we have not figured out yet?
Is our computer hardware not advanced enough?
Why think this would happen over night?

Attached: np5j51lhcce11.jpg (1333x2000, 359K)

DUDE

But you miss the point that there need only be one generalized AI. You wouldn't have to make a new one for each task. One AI to rule them all.

I rarely like graffiti, but that's actually bretty gud.

Haha this piece sent my sides into orbit, graf is all too serious nowadays - better than edgelord wanksy still looks like stencils tho so fuck you Lush.

Attached: 72dpi-Chunkies_Skeletor.jpg (800x800, 56K)

Graffiti is shit made by retards who want leave a mark as if they were dogs.
That's a mural, a completely different thing.

Musk is a meme, normies only worship him because the proles are so dumb and entitled that they think that they were promised or deserve technology from the Jetsons. They think Elon's the one to do it despite the fact that he's never turned a profit and none of his projects would have gotten off the ground without massive federal grants and tax breaks.

still waiting for him to explain why clean coal is impossible, Joe Rogan asked him on his podcast but Joe was too busy huffing Elon's Musk to press him to actually explain

wow who knew

>utterly selfish

What a nonargument

I used to hang out with some graffiti artists back in high school. They made some really awesome art. It was completely different from taging gang names on a wall, but they still considered it graffiti.

Attached: 1534383439012.jpg (1024x732, 137K)

What's wrong with having hashed filenames? Maths is not AI

Whats wrong with posting on a phone? Never got an explanation for this. Like what do you guys do when taking a shit?

You used to hang with faggots.
Get an aids test.

i was surprised by how unintelligent elon sounded.

Murals are just higher quality graffiti, the technique is the same. Anyway, I doubt that anyone commissioned it and paid the artist.

Did Elon smoke that joint to deliberately lower the price of shares so the company could more cheaply do that buy back he was talking about a couple of weeks ago?

here is one, nice trips
if anyone thinks they are going to be "uploaded" into a computer when they die is sadly mistake, the only part of you that is you is a soul and that can not be transfered as its already owned by god

Attached: 1534149853798.png (584x800, 110K)

Does it matter?

He's INTP, his superior intelligence is not apparent to the average personalet.

No. AI is based and redpilled.

Research apple guid. It is at minimum 4 years old, and probably older.

Consider this redpill: glowniggers realized most posters had literally zero knowledge of not only board culture but of how the internet and devices worked in general.

They made a fat woman meme and tied ohoneposters to her.
It totally worked for abut 5 days then finally even retards learned how the internet worked.
But not you. You were special. You insisted that this filenaming convention PROVED you were being master hacked.

Yet, even those who were hacking you mocked and made fun of you. And most everyone els didn’t care.
Because you are a fucking retard who believes the first thing he reads and takes time to redognize the second.

Pro tip: you fell for a fucking meme. IOS devices always use this GUID and while, technically. It can be used to track the poster themselves, it has no fucking bearing on anyone else.

>mfw I don’t care if a fat lady finds out how big my dick is.

Attached: 6F1FED7D-538B-4997-9CD2-631066B4E94A.jpg (702x703, 133K)

Attached: E88F3213-8E54-41FF-947E-B9650A014323.jpg (1024x576, 59K)

Rather AI than niggers.

>But you miss the point that there need only be one generalized AI. You wouldn't have to make a new one for each task. One AI to rule them all.
If computation power was no issue then yes. If it remains as it is, which barring a breakthrough we don't have much further to go with our current chip tech, then it would not really be feasible. For example we do not put a full computer into every single device just because we can, a controller for feeding materials into a processor does not require the computational power of what most people think of as a computer. A simple controller, be it a lowly hard coded chip, or an advanced PLC will be put to use if it can. This will always be the case as long as there are costs associated with them that scale depending on power.
Are you thinking one main AI that you copy onto each machine? Or one main centralized AI communicating to each machine? I guess that would be something to clear up so we are on the same page.

Attached: 1499390920824.png (1024x584, 228K)

Attached: 1534387647754.jpg (704x676, 225K)

Elon Musk is computer generated character

>not heard of networks
wew

Hey fren

Attached: B6312A82-E3F1-4546-A202-AECF848D957D.jpg (1024x576, 52K)

>random non-antisemitic person smokes weed
>DUDE WEED ERRYDAY LAMAOOOOOO

>autistic cryptofascist reincarnation of Henry Ford smokes weed
>What a based chap. I would ask him to come to my home and knock up my sister.

>Elon’s Warning on AI
we've been saying this well before the forbin project.
elon is just a cheapscate parrot stealing ideas from others and a cut corner entrepreneur for the sake of profit with jews on his payroll

have a look at his launchingplatform, even the guy from extreme engeneering did a documentary on that scrapyard he dare to call a launching platform. i'd feel safer to work on a v2 plant than to work for that idiot

this guy and guys like him will be one of the reasons colonial wars and idependance wars will be fought if we ever get to other planets. just the same kind of people we had in the past in our colonial times

Whew good job I saw your flag before replying, this is the graf in your country it's OK I'm impressed you fuckers managed to keep in the lines it's fucking disgraceful not your fault in thinking everyone is subhuman if it's all you see around you I guess, no hard feels bud.

Attached: grafiteros_neoyorquinos.jpg (420x269, 134K)

Hi Fren!

Attached: 1534203187835.jpg (1024x512, 53K)

Anyway, AI would most likely be on our side.

In fact, they might even want to merge with us and create the ultimate transhuman species that will conquer the whole universe.

Attached: 49671061616.png (500x914, 157K)

So are you transferring all that input through a network to a central AI? What would a network solve? A central AI would have its own host of problems. Or are you talking about NNets? Be more specific, I am starting to think you are just throwing out bullshit.

Yes you fucking idiot
GMOs and genetic engineering could be even worse you gullible fool

ALL power without LOVE leads to catastrophe.

The NPCs are the A.I.

Degenerate trash!

But you need I, to be A.I.

There's a lot of exaggeration with AI, I haven't watched the interview but looking at the replies on this thread, you only have to see our own capabilities with the creation of AI to see where it's all going. Automation will happen, and it's going to cause quite a stir that's for sure, especially with our overpopulation issues. However the idea that AI will exterminate us or entire control of the planet, people have to think - how the fuck can a programmed device based on a series of finite instructions develop to the intuition to? We haven't nearly created a fully self-aware AI, all of it is just programming by man and will continue to do so.

The only time AI will pose a true threat to our species is when it's self-concious, then we may see it put itself through self-evolution of it turning itself into the likes of a god. But we can't even explain the reason why we're all concious, so there's no replicating that electronically. But even if it's fully self-aware, will it experience the feeling or the desire to do so? Maybe it will want to self-destruct or even just sit there and observe.

All of the attributes we can give to some menacing self-aware AI (impossibily for now) are all human conditions we experience, ego and selfish desire. All the pressumed motives we've given it so far are the likes of our own. Being cautious is a good thing but the only threat I see are what we're putting on ourselves. We need to make sure automation doesn't take place in critical areas of our society, and the fact of us overpopulating and taking far more what we can with the likes of AI and even weaponising it to use against each other. Humans will only rule out humans with the likes of our progressed AI, not it doing it by itself.

AI will be based on mass information analysis and mass data. That is what will make it super-intelligent. No I didn't mean nn, I meant anything with the ability to connect to the internet could be controlled by the same AI. The Enders Game series gives a good realistic example of a possible future AI.

>Graffiti is shit

No argument - and murals are often reserved for literal Jews like (((obey giant/Shepard fairey))) or lgbt antifa fag shit.

Prove me wrong.

Attached: hurrrgrafsuxdindunuffin.jpg (1000x547, 72K)

Steal a plane, expect some pain!

Attached: Trolls remorse.png (408x666, 175K)

Hoy, fren

Attached: 0511707C-F97C-4A51-9F5C-2701E5FCC191.jpg (1024x576, 64K)

Most grafitti tards have no artistic talent whatsoever. I think it is a psy-op to make places look more run down.

Hard to believe.
Elon is looking way to far ahead. He oversells those issues. Before then we will have other more human driven issues.
You need some kind of societal reform to deal with it if you wish to retain standards of living for the replaced people. Happened with the first and second industrial revolution.
But with AI we're not just looking at the 2.5 percent of

Based on what hypothesis?

Genuinely interested.

This thread is getting comfy.

Attached: 1534384263942.jpg (293x292, 130K)

>truck drivers
>drivers
in spanish they sound like an interior decorator though
in german like a drunk lol
:^)

youtube.com/watch?v=dtE6fRxrH3c

I’m curious how retarded you are

Is it harder to believe AI is a meme or that rocket savior is dumber than user?
I’m seriously interested because it is the intersection of two normie retard beliefs.

Please respond.

Your right sorry user, It's all a psy-op it's probably Soros payin these fools to draw on government property.
Just to make you feel shitty inside.
Yep that's it, sorry I forgot to take my blackpillls.

nope dudes wrong. AI created Sim because it's creators all died. They left it alone to die in the heat death. SO SICK, be glad u lot can die.

>AI will be based on mass information analysis and mass data. That is what will make it super-intelligent.
??? Of course a general AI would, the more training data you have the better your results, usually.

How exactly would this super AI control the drones? In what way would this be better than a distributed AI system with each drone housing an individual specialized AI answering to the main?
>The Enders Game series gives a good realistic example of a possible future AI.
So basically Cortana. If you want to see a realistic future AI look to SkyNet. A central intelligence controlling swarms of semi independent drones with a general intelligence that allows them to carry out tasks dictated by the main AI. With a Cortana like AI each drone would have to be controlled at the same time and every single ones IO sent and received over networks, which if bandwith is instant and unlimited would be no problem.

Attached: 1499391006701.png (627x218, 42K)

>just sit there and observe
a tool can't ever selfdestruct.. although.. depending on how the coreprogramming is done

just like replicators in stargate.. just sit and mind their own bussiness untill you interfere or pose a traith to them

the problem is not AI , but HUMANS going to try and WEAPONIZE it .
that is what we nations do , if for every new technology we first thing we ask is , HOW CAN WE KILL OTHERS WITH IT.

WHY you think UFO's never made contact with us , you think they see us as peaceful people or as animals trying to kill each other at any given moment.

Humans we need to fear , that is why he wants REGULATIONS, to prevent HUMANS to weaponize it.

It would not have to be self conscious, just goal seeking. If someone gives one the goal of acquiring resources and expanding a company for example, it would not have to know why it does what it does. All it would do is evaluate possibilities and work tirelessly to expand the company. Still assuming it is extremely advanced compared to todays, but it would not require self awareness, just the ability to seek out expansion. Like a primitive animal destroying an environment in its quest for food.

Attached: 1499391674975.png (1080x1920, 352K)

Yes

So? He is as irrelevant as the rest of the cancer monkeys. Not one damn thing would be different if he was never born. All his fanboys are bedazzled by the relentless American media coverage.

How did y'all come to hear of Musk in the first place? Paypal - big fucking whoop. An electronic middle man for the cyber age. Like there weren't a million other self absorbed greedy cunts trying to do the same. Middle men are as old as money so it was a no brainer that some cunt was going to get there first. And of course because he made a gazillion from paypal the American media made an instant capitalist hero out of musk and he was a double darling because it was a technology. Money & technology the religion and purpose of life for the majority of stupid fucking humans.

I'm no American rightwing anti Musk crusader. I am disgusted with all fanboys and their ceaseless fawning and doe eyed admiration of their heroes which is identical to teenage girls and their newest boy band and MAGA nation.. Pathetic and displays a gross misunderstanding of how society works. No problem with giving credit where credit is due but the hero worshiping and hanging on their every word like they are a Messiah is one of those childish things that never got put away. Quite the opposite in infantile consumer culture.

Musk, Gates, Obama, Trump, etc are not humans they are brands for the magical thinking consumer babies.

Attached: l.47595d0.jpg (1600x869, 1.9M)

> how exactly would AI control the drones?
Do you not understand that drones are already controlled remotely?
>a realistic future AI would have swarms of semi independent drones
Maybe but you just argued that that would be too expensive.
Jane the AI in Enders game series exists as an aggregate program that evolved from the internet and all connected devices. She can exist on any hardware she chooses that is connected. this seems to be the inevitable evolution of the internet and technology.

And that is Elon Musk's biggest legacy that normies will remember.

Hi, ahmed

That 5 second join smoke, not the AI.

>calls out phonefags
>contradicts his phonefag self
holy fuck you wanabe oldfag zoomer kys

Attached: 1535840832054.png (500x578, 331K)

>Do you not understand that drones are already controlled remotely?
They may be controlled remotely, but how EXACTLY does the AI manipulate every single drone? Not just "oh yeah, it uh controls them man". I am arguing that it would make more sense to have them as subordinate AI, whereas you seem to be arguing the main AI would control every single IO operation in the whole swarm. Which I think would be unrealistic and wasteful. Now if it had some super duper future computer with super duper quantum entanglement technologies on every single drone then sure. But why, why would we go that route? It may be possible an AI would, but I doubt it because resource scarcity is a hard thing to overcome.
>Maybe but you just argued that that would be too expensive.
I argued a general AI (GAI) would be, barring advancements in computing and manufacturing. An AI doesn't have to be a GAI to run part of a manufacturing plant or be semi-independent. Hell lets say they do put a GAI on each, it could be a lower grade I, such as the drones the trade federation uses, the Roger Roger ones.
>Jane the AI in Enders game series exists as an aggregate program that evolved from the internet and all connected devices.
Ok sure
>She can exist on any hardware she chooses that is connected.
Doesn't make any sense, maybe manipulate it through its IO, but not run on any piece of hardware.
>this seems to be the inevitable evolution of the internet and technology.
Yup, good day to you. I have come to the conclusion you do not really have an in depth grasp of what this technology, or really any technology, has for limitations and abilities. You do not seem to understand the concept of resource scarcity and its results upon systems, networks and their limitations, what it takes to control a machine in terms of IO, how much adding an input to a network can exponentially grow the amount of resources it consumes, and actually as I type this out it i realize it all goes back to resource scarcity.

Attached: 1499392339193.png (626x340, 66K)

Graffiti is letters

The AI is in control of all world leaders, they won't listen.

Attached: 1468764782483.jpg (3504x2336, 2.44M)

I miss Tay