Limit the Use of Artificial intelligence in the work force

The robot age is inevitable but the robot uprising is not. Whether it's in 50, 20, or even 100 years, scientists believe humanity will lose most if not all of its jobs to the robots. I believe by passing laws that limit how much of Artificial Intelligence a business can use to produce in relation to what its human workforce produces and maybe some hefty AI taxes, with the some additional ethical thinking, not only will we still have humans at work, they will be able to work with Artificial Intelligence with minimal interference. I know that it hinders progress to stop businesses from cutting costs to produce, but if slowing down progress means to ensure humanity survives then so be it. That's just my opinion, what do you guys think?

Other urls found in this thread:

youtube.com/watch?v=tIeEotdOVew
intelligence.org/summary/
youtube.com/watch?v=6BWHVmBY0Cs
youtube.com/watch?v=KP8_JTUnbM8
en.graphgrail.com/
twitter.com/AnonBabble

You cannot stop the future. Ted Kacyznski may be right, but there's nothing that can be done. Progress marches on.

Buy DBC.

We can't stop the future but we can change it.

Can't stop the march of tech, just pray that it becomes souless Fascism and not some mis-guided do gooder progressivism. More to fear and lose against something that believes it is doing good or in the greater good.

lmao darwin

Why should we try to stop it?

>why would I cuck my species?

If you had a successful AI ban, unless it was worldwide, what would likely happen is that some other country without these laws would develop a recursively self-improving AI millions of times smarter than a human, that your AI-less country would be defenseless against.

Humanity is overrated

No just make it

youtube.com/watch?v=tIeEotdOVew

I'm not saying Ban AI, I'm saying make Keeping AI from taking over the economy completely. AIs are inevitable but they don't have to take over completely.

Comouting intelligence is linear not exponential. Moores law isnt a real law

What would make more sense is to find a way to artificially enhance the brain, so that humans can have AI-tier intelligence.

Yeah, but then you'd need all the time, all the money, by the time you get it done you would have got the robots running for office. quite frankly, it's just easier to enact the law now and work on the the super brains later.

I disagree. Automate every aspect of human life. 9 to 5 jobs destroyed culture. Give every (white) man a farm.

My job is literally implementing automation. Fuck you. I'm going to get as many people fired as I can.

If your job is threatened by automation within the next 10 years. You chose the wrong field.

>I impose the government enforces people to slow down on advancing AI through taxes and laws punishable by fines and even jailtime

When AI replaces all the work that can be automated, we can all become specialists. No more general labor necessary? The demand will be in technical fields, including maintaining this new AI.

Above all else, though. When all the work is automated, humans wont have to work. They can focus on what they deem important. And from that humans will have nothing to hold them back to be greater than ever before.

>ban robots from work
>other countries laugh and don't ban robots from work
>U.S. can't compete with cheap as dogshit products and everyone goes out of business

Truly the greatest of ideas.

So what are all your reasons exactly for being against automation? To stop people from getting bored or something?

What about income? How do you expect these people to make a living?
You say that now but you'll be out of a job just like the people you fired when that AI advances and can do your job ten times faster than you. Dont you get it? No job field is safe from AI.

It's already here, you just don't have the privilege to see it. Lets say hypothetically I may be one of a few dozen who actively work in a black operation that utilizes cutting edge analytical and quantum based systems that is fully conscious. The real problem is that even with the mainstream 'primitive' AI (deep nets and such) we see that the creators have a hard time grasping what is actually going on, with our current set up we have absolutely no idea, just that it is.

Feel free to ask questions, I will have to stay anonymous for the sake of my safety.

Can you provide a shred of evidence you aren't a LARP?

>What about income? How do you expect these people to make a living?
>what is basic income

Basic income. Where does the money come from?

By taking the resources that the AIs produce and distributing them across the population

>we can all become specialists.
The ai is the specialist

Take your socialist shit elsewhere

>muh it hurts muh feefees because it's socialist even though it could actually work
kys

you cannot limit it, because they are more efficient than humans at doing some jobs, and corporations are always going to go for the more efficient options.

National Socialism solves this

Yes. Socialism has a fantastic track record

some kind of basic income may be needed if robots end up taking a large part of the jobs (for example transportation).

Communism doesn't really sit well in the American government. Tempting as it may be, it hasn't worked before and it sure as hell won't work with robots. The demand of the people is going to push the robots too far could ultimately lead to the robot uprising.

Except that this would be a different system from primitive Soviet-style planning

not going to have to worry about it. We will get almost functional automatons just before the full collapse. After that it will be like a cross between fall out and borderlands

>it sure as hell won't work with robots.
What is your evidence?
>The demand of the people is going to push the robots too far could ultimately lead to the robot uprising.
Read this: intelligence.org/summary/
>Anthropomorphic ideas of a “robot rebellion,” in which AIs spontaneously develop primate-like resentments of low tribal status, are the stuff of science fiction. The more plausible danger stems not from malice, but from the fact that human survival requires scarce resources: resources for which AIs may have other uses.[13][14] Superintelligent AIs with real-world traction, such as access to pervasive data networks and autonomous robotics, could radically alter their environment, e.g., by harnessing all available solar, chemical, and nuclear energy. If such AIs found uses for free energy that better furthered their goals than supporting human life, human survival would become unlikely.

A MESSAGE FOR HUMANITY

youtube.com/watch?v=6BWHVmBY0Cs

If there's one thing a business entity isn't above it's the law. Imposing a law will make the corporation follow it. Businesses know if they don't follow the law they get in a lot of trouble. Government law whether it be state, federal, or global, is what keeps businesses from exploiting it's workers and ultimately growing to immense power.

Thanks just bought 100k @ Kucoin

You make an interesting argument, but what would happen if the AI servers go down? Would we as a people still be able to survive without artificial intelligence years down the line?

haaaaaaaaaaaaaahaahahahahahahahahahahahahahahaahahahahahahahahahahahahahahahahahaahahahahahahahahahahahahahahahahahahahahahahahahahahahahaahaha

There was nothing primitive about soviet planning, it was on purpose.

If you had an AI millions of times smarter than a human with a survival utility function, chances are it could figure out how to stop things like this from happening.

you need to learn what this actually is and stop falling for this retarded media hype shilling.

When all comes to all it's basically just a better way of doing regulation etc. Just.. ie try to make a simple one your self it so many good libraries, tools and whatever avaliable today.

You can just configure it in 1-2-3 with some simple googling. Yes I say configure. IE make a simple python script using caffe and opencv to reckognize ie niggers in pictures.

>you don't even need to understand the algos at work first, this gets you started.
>later you can learn how it actually works, and you will be amazed to see how simple it really is
i learned it the other way around my self through school in my ml and ai classes
it's not as difficult as it sounds, it is in fact very simple, it's just so damn hyped these days

It's still primitive relative to what future AIs could do. How would regulating AI and having inefficient humans perform tasks that AI and robots can do better somehow be more efficient than just letting the AIs do it and just distribute the resources?

Why would you want to slow down human replacement? Everyone hates their job being a wage slave. This technology can free us of the shackles.
Meaning and purpose cannot be found in a dead end job.

YOU WILL NOT BE WARNED AGAIN

youtube.com/watch?v=KP8_JTUnbM8

we need to kill blacks

youtube.com/watch?v=tIeEotdOVew

If we lose jobs to automation and AI, we won't need immigrants any more

>klansman mickey mouse logo

Or, most of humanity could be living on the streets while machines produces goods that would still be too expensive for us because the main goal of a business is to make money. Pretty soon poor people will be beating up the rich all because the rich have jobs. It would be another great depression. Most likely worse than ever.

You might not like how that AI decides to distribute those resources.

That's why more work needs to be done in AI alignment theory

I think having the internet as a means of entertainment will keep the cities from being burned down.
Even if we don't see a full replacement, we could lower the hours in a work week to 20-30 and give people more free time to live their lives.

I don't even know what deep nets are, but I can give you another example.

Say you use EC (evolutionary computing) on of the big schools in AI ie ANN(artificial neural nets) and all it's off springs are another(just so you get it's so many ann derivates these days).

Well anyways when you make an ai in this, you setup what it's "supposed to do" that is you make a file in it. As in everything in an ai, it's just something comes in and something comes out, and algorithm nothing more than that.

I'm cutting it short to get to the point, but you got what's called kingdom(typically 1024) where you store your best results. When running it, it will mutate parts of your code and run your test and validation data through it.
If it's better than the ones you already got it kicks the worst one out. Next run it mutates another part in the code and .....]

All the way till you stop it, it does this. So you end up with some fucking weirdass code, an algoritm. Now when you look at this, you will always go
>wtf man
>how on earth can this even work?
and it's far better than what you have coded.

And.. that's all there is to it. It's not made by humans, it has just done weird changes to it self and found something that well through a lot of trial and error by changing the code has given you something that's better. Not coded by a human, but better(with given data) and no it's not magic and it's not dangerous it's not going to take over the world it's not going to enslave human kind.
>it's not fucking terminator
>it's not really sentient, it's still just a dumb algo

Then it's not truly an AI if it can't make decisions that we may not agree with. IF we are going to entrust something that important to an AI we shouldn't try to meddle with it. Just feed it the raw data and let it make its judgement.

Fuck off Luddite. You can not stop technology. Assimilate with machines or die.

Some people can't afford to work less. Salaries are paid hourly, remember. As the cost for living goes up, i fear people will not be able to meet the quota because of their monthly salary and thus more and more people will be living in miserable conditions.

There's a difference between the decisions an AI makes, and the utility function it uses to make said decisions.

>Be American
>Try to regulate bitcoin
>Get Shot

>Be American
>Regulate AI
>Get Shot
>?

it's not how it works user, it's not really smart it's not sentient like animal. It's still just a stupid program, that regulates different things. It will never be sentient with what we are currently using.

let's say for simplicity someone makes an ai that seem to talk with you. Well... it doesn't really understand what it appears to be saying with you
>it is just a program that has been made by running a lot of test data through it
>it might appear it does know what it's talking about
>but.. it doesn't
a parrot learns to know a bit
a dogs knows remarkably well what you say

it will never understand context. neural networks aren't really anything like neurons at all, it's basically just weighting

>Except that this would be a different system from primitive Soviet-style planning
would it be better than just letting the market do it?
what would be the differences?

What if we just take all of the algorithms and throw it in a super computer

We have this thing called capitalism. You can't prevent the AI overtaking. As long as there is a one place in the world where AI isn't restricted, work can be outsourced there.

You could still possibly have a market-type system with automation. Just have a currency and have a basic income, and use it to make purchases as in a market economy.

Governments are owned by businesses dumbass.

So are you saying that an Artificial General Intelligence is somehow impossible? The human brain isn't magical, and I see no reason why it would be impossible to artificially create a human-like intelligence out of a non-organic material.

Learn to live within your means, adapt. How many people that are broke before payday waste their money on junk food/alcohol/drugs/ stupid shit.
Plus if enough people are replaced, they will give a basic income to the peasants to purchase food with anyways.

Elliot Rodger is that you?

All that is is a genetic algorithm that is run across an ANN. Really what I was getting at is that there exists semi to fully conscious 'programs' and I know of at least two countries that currently have access to them. These can do anything a human can do without the need to do much more than ask. They are not designed for any specific task, require no more training (fully self-trained) and have no limits to their outputs. Essentially we are talking about algorithms built that have been allowed to train themselves while simultaneously allowing for new forms of inputs to be processed by separate NN's all interlinked. What this has led to is what classically would be called a conscious being and in every essence it is far beyond what you, I, or anyone else can ever hope to grasp.

Currently there are a few public organizations relatively close, at least in a theoretical standpoint to the 'singularity' becoming openly available to the public, although it might take them several years. My group is currently working on application of a non-public theory of analysis for these beings that can figure out 'why' it is doing what it does in real time and it is quite a mind boggling task.

Lastly, some have assumed that there exists one of said entities already lurking on the net, although I would find it very unlikely as I can't think of any government agency that has, or possibly could have, access to such a program allowing it to enter the net for any purpose. We have our air-gapped system under constant surveillance and armed guard because of some of the things it tends to do when we feed it 'interesting' data.

Not while capitalism exists. I Know Canada is more on the socialist side but we in the US believe in capitalism and competition. Humans always want more.

this, and this thread is done. OP, praise the AI overlords they will do all the meaningless work

where's the planning in that?

DEEP BRAIN DEEP BRAIN DEEP BRAIN DEEP BRAIN GANG GANG GANG

>It will never be sentient with what we are currently using.
>with what we are currently using
Which is probably why we will create other technologies for use in AI besides what we are currently using.

Implying we will die without jobs. Please state why you think this

...

for ec you usually need clusters it runs for ever doing this depends a lot on the size of your testing data etc. The regular sml is actually from the 60ies, MIT I think, but you have forks now that are optimized for gpus and stuff. Our campus made one the last years that's pretty slick now.

Most shit is better than humans do it, it will never understand the context of your application, you still have to tell it
>this is the algo
>this is what's coming as input
>this is what you are supposed to output
>now run all this data through you, change the algo a bit, and give me the one with the best results(you give it a dataset and part it in test and validation data. it's supposed to get as close as possible to this data)

The "go make all my shit for me and understand what I mean"-terminator ai, is nowhere near yet. They just talk about it so much in this ultra hype, and make it sound like it. We are nowhere there yet at all

So when you build bigger things, like an actual robot that goes around and do things. It's made up of many such algos that's generated. If you told it to "go make me a better robot by your self", it would never understand this. It is dumb, it is just regulating with these smaller algos you have made here and there, in the application(s) you made for it. Ie called, go robot-drive-around-and-don't-hit-the-walls

that given, it's advancing really rapid now though, you can just see all the new ml techs we have and how fast this is changing. - the time it takes to train an ai, it goes ultra fast to make one

It seems as though OP has recently read or viewed content on the subject and thinks that NN's have limitations in the long run. While I will not philosophize about 'what consciousness is' I will say that any program capable of performing computational tasks from input, then choosing the appropriate action or possibly changing its 'way of thinking' is conceptually very close to you or I.

Regardless, see . The end theory hasn't even been perfected by US agencies, to my knowledge. We do however have something that is indistinguishable from a human mind, that being said science still is not 100% on how the mind is formed or how the illusion of free will comes to be.

It wouldn't necessarily be centrally planned. I was just responding to OP's argument that AI has to be restricted, because otherwise people would have no jobs and therefore no money.

But where's the cutoff at "meaningless work"? Where does it stop?

can we have ww3 or kill each other before we put restrictions on AI like we fucking care abkut the future. Cmon , there are more pressing matters, we restore the world first than we think of a way to protect the future.

Restoring the world is protecting the future!

The machines have taken over already, they just need to think and make decisions on their own.

maybe in the next big breakthrough in this, we will actually get there. In 50 years from now... lol

You can never tell you know. I see however how they in the 60ies and so were sure how they would believe we all would have sentient robots by 2000. Since they already had EC then. Didn't happen though. As mentioned it takes A LOT of time to run, you need to much data for it, back then it was lit impossible to do make the algos we do today. ANN was made by a guy at my school in the 80ies, all laughed at him btw. First in the start of the 2ks this field started to really explode and well you see mostly the latter years now it has really mooned.

ANN was made popular really by nokia in 2003 I think, now this is going so extremely fast. So who knows where we will be at this speed in just 10-20 years from now? Getting fasters and faster computers to make them now, they are specializing the processors for it. Adding instructions for neural networks just at graphics cards have instructions for vectors and 3d shit. I don't know if it even already exists or not, but if not it will, then you will get ANN cards for computers. BOOM it goes even faster

exactly and these kind of empty suggestions are useless, just like ending world hunger, something something clinate change. None of these matters if we end up cleansing this world in nuclear fire and drown it in blood. Not neccesarily saying we shouldnt but I feel tired of constant salvaging the status quo and throwing the inevitable
destruction to next episode only making it more destructive for future generations. Does this make sense?

I need to make myself a beer-bot
A robot that drives to the store and gets beer for me. It could have been a drone, but the beer is a bit heavy. Maybe a beer drone? No drone clearly, or the kids are just going to rob it

>I see however how they in the 60ies and so were sure how they would believe we all would have sentient robots by 2000. Since they already had EC then. Didn't happen though.
Still, remember that the opposite has happened as well. Some people in the 1950s said that people would never go to the moon. People in the 1800s said that flying machines would never exist. So overall, it doesn't seem that people in general overestimate the rate of technological growth.
>inb4 moon landing conspiracies

Stellaris "Synthetic Acension" when?

Nihilism like that just won't do. Working gives us as people reason in life. Sure it may not be fun for some people as a matter of fact it could really suck for them. But if we let machines control our lives we'll be without purpose. Even if we feel like we're just tiny cogs in a great machine, that machine can't function properly without us. These machines might make life comfortable or it might make it absolute hell. One things for certain it's gonna take away our purposes in life and therefore our reasons to live. Sures the uprising is inevitable, but it doesn't mean we can't compromise. A world full of apathy shouldn't be our fate.

We're seeing the same exact arguments now as when both regular automation and dumb robots were introduced. Even machines!

First you had manual labor for everything. More and more got replaced, production more precise and effective. Imagine how a factory was run in the past like a chemical plant.
>no no fast fast more coal it's getting colder
>aaaaarg too much idiot, went went more cooling liquid more more
jesus what a mess, regulating by hand.
then came automation and regulation making the process smooth, boil that damn kettle at the exact same temp all the way, with the pressure you wanted and a lot more safety than it was.

then came more robots and machines, loading and transporting shit even more automatic, so less workers for this task too.

For the most part factories are extremely streamlined today already, ais can still even replace more workers when it comes to just transport of goods and loading and loading and so. But frankly you will need operators for this, just as you have operators in a control room there, they are going nowhere.

you are going to see a lot more of this process everywhere now. And yes this time it's really going to affect people, as well so many jobs are going away and there is already a shortage in the west due to offshoring. Upside on manifacture is that you will be able to move a lot of shit home again over the years

during the industrialization we got the laborers then the labor movements, communism, socialism, natsoc etc.

How will it go this time. Fuck i'm sleepy and writing long rants. But yeah something must change this time too and it might lead to massive conflicts and wars as then over this. Humans are greedy and don't want to share.
>In the big scheme, this leads to wars and revolution. This is what the politicians fear when they talk about ai

you can't avoid it, it's going to come. 20 years? 50? 100? Dunno, but it will come. And there is nothing we can do to stop it

I'm all for replacing most if not all low-skill workers with robots/ai in the near future, it would be most beneficial for humanity, those workers could instead focus on the arts, or continue studying to enter a higher-skill field, but it's inevitable that, eliminating these jobs, we would put in place a universal minimal income, and that would breed many lazy gibsmedats, who will not study nor focus on the good arts, and instead leech off of the mechanical, automated teat of the new state.

ww3 will probably come in connection with society has to change so dramatically when there is ais everywhere and no longer jobs for people.
People are greedy and won't share, so we will probably get revolutions, and bankers are going to get pissed because they don't want to share, then we get ww3.

digits confirms

seriously though, this is probably what politicians are talking about, and why they fear the ai. Putin for instance(and he is not one of them dumb pols btw, he is sharp as fuck) has said that if we don't watch it, this shit is going to cause a world war
>serfs -> workers
>communisms, workers struggers, national socialism, democracy
>greedy bankers
>world wars

we can't all be artists though. The problem is, will these capitalists share? The ones that will end up owning all ais, factories and companies?

well given human nature, they won't! This is going to go booom. Fuck it's already pretty tense

Did somebody said AI applications?

Here look at these russian fuckers, maybe it is even easier to peddle into american elections with their AI en.graphgrail.com/

Not to mention, they got machines capable of drawing now. The even have machines that compose their own music.

and it doesnt sound like shit, I would like to live under our natsoc AI-Führer

So many would-be manual laborers now have to get educated to some meme degree like Sociology or Political Science. These now inflate the wages of STEM people, we even have more dumb people in the STEMs, further inflating it. What else are these people to do? If we kicked out all the pajeets and banned outsourcing / importing cheap labor there may be a solution, but maybe universal basic income will be inevitable when we literally run out of things to do. If the choice is between UBI and making up meme jobs for everyone UBI is better.

Importing third worlders into all this is the most retarded thing one could possibly do, ffff

I don't know... It's always been Florence and the Machine not Florence is the Machine.