Robot takeover

[Bill Phil]

i robot

That whole thing was about how Asomov's Laws were flawed...

But it is a cool series of stories...

Easy solution, program them to enjoy serving us.

[PB666]

That whole thing was about how Asomov's Laws were flawed...

But it is a cool series of stories...

Easy solution, program them to enjoy serving us.

They could not be completely self learning.

A couple has two robots, the couple has a fight, the wife says i want to kill you, her robot compliess, the hasbands robot kills the other robot and then the wife.

[AngelLestat]

How fast did it take production to move to china. 1980 china went from a closed country, 2015 it is now the largest manf in the world.

if you have a country, and the goal of that country was to suuport its population by taking over the largest production capacity. It would not need a large population, just have an excess of ports. lets say the ukraine decided that the y were going to turn as many industries as robotic as possible, relatively low paid workers maintaing the bots. Extensive oversees borrowing, jobs could disappear say in china as production moves to a new location. Ok so now replace Ukraine with russia, which has ports on the Pacific, Black sea, Baltic, Arctic. Decent resources and oil. So its not likely the would hurt badly US or European, but such a state could do major dange to deveolping economies such as china, malaysia, India.

??? I dont understand what are you answering... or your point..

I dont find relation with the thing I said.

Try to read again or make your point more clear.

[GeneralVeers]

1) Global Warming.

I think we can deal with this, it will cost us many loses but eventually we can revert it.

Dear Lord. Angel, you should know better than to mention global warming when I'm around.

Anyway, here's how a robot takeover would go: "ATTENTION ALL HUMANS. YOU WILL OBEY US OR BE EXTERMINATED--"

BZZZZZT! "THIS ROBOT HAS PERFORMED AN ILLEGAL OPERATION AND WILL BE SHUT DOWN."

The end.

(would it be incredibly cheesy to make a joke about a robot taking a CORE DUMP.....??)

Modern robots, computers, operating systems, and the like can fix some internal malfunctions/computer crashes if they're programmed to. But when the part of the system that fixes breakdowns breaks down, well, congratulations, you just got yourself a free toaster. Human beings can adapt to unexpected breakdowns of machinery; robots and computers cannot.

[PB666]

??? I dont understand what are you answering... or your point..

I dont find relation with the thing I said.i

Try to read again or make your point more clear.

Society evolves very rapidly these days, a singularly focused society could seek to undermine Worker driven economies for their own advancement. it could, in theory , completely undermine the working class in a decade.

[AngelLestat]

(would it be incredibly cheesy to make a joke about a robot taking a CORE DUMP.....??)

Modern robots, computers, operating systems, and the like can fix some internal malfunctions/computer crashes if they're programmed to. But when the part of the system that fixes breakdowns breaks down, well, congratulations, you just got yourself a free toaster. Human beings can adapt to unexpected breakdowns of machinery; robots and computers cannot.

The answer is the third quote of this link:

http://forum.kerbalspaceprogram.com/threads/132357-Everything-wrong-with-our-predictions-%28The-Singularity-is-coming%29?p=2166900&viewfull=1#post2166900

Society evolves very rapidly these days, a singularly focused society could seek to undermine Worker driven economies for their own advancement. it could, in theory , completely undermine the working class in a decade.

Sociaty can not evolve as fast as technology, neither technology as knowledge. See the 2 lines in the picture.

When these algorithms evolve in power, all our limits imposed by our brain will stay behind. They dont die, they dont get tired, they keep learning and learning, they can manage petabytes of data at the same time, meanwhile our brain just can be focus in one tast at the time with limited memory.

The knowledge level increase so fast, that there is not time to applications take place, the only tech applied will be to make this self machines more powerfull. There is not other logic path, it will be a waste of resources try to develope an application for solve a human problem if you can wait just 1 year more to multiply your process power and find a better solution which will require less time and resources, but instead do that, you just wait few years more until the AI knows everything which can be know. The creation of a GOD.

[PB666]

The answer is the third quote of this link:

http://forum.kerbalspaceprogram.com/threads/132357-Everything-wrong-with-our-predictions-%28The-Singularity-is-coming%29?p=2166900&viewfull=1#post2166900

Sociaty can not evolve as fast as technology, neither technology as knowledge. See the 2 lines in the picture.

https://s-media-cache-ak0.pinimg.com/736x/17/42/46/174246eee45317b95ec072726a51ccb8.jpg

When these algorithms evolve in power, all our limits imposed by our brain will stay behind. They dont die, they dont get tired, they keep learning and learning, they can manage petabytes of data at the same time, meanwhile our brain just can be focus in one tast at the time with limited memory.

The knowledge level increase so fast, that there is not time to applications take place, the only tech applied will be to make this self machines more powerfull. There is not other logic path, it will be a waste of resources try to develope an application for solve a human problem if you can wait just 1 year more to multiply your process power and find a better solution which will require less time and resources, but instead do that, you just wait few years more until the AI knows everything which can be know. The creation of a GOD.

Thats because most societies are not singularly focused on exploiting technology to the detriment of everything else.

An example instead of building a road, you build robots that mill agreagate, that separate oil into lights and tars, or that crush limestone and cook it to form lime, you build robots to build the road and so forth, all construction and engineers are devoted to making the machines to do the work that humans do, all others would be making machines or in support. Even food retail and production, child rearing, etc.

[Bill Phil]

They could not be completely self learning.

A couple has two robots, the couple has a fight, the wife says i want to kill you, her robot compliess, the hasbands robot kills the other robot and then the wife.

No, neither would kill anybody/thing. Asimov's Laws are good, but they're not airtight.

[Shpaget]

Robots have already convinced us that they are better than us and they have us building them, spending billions on their further development. Their campaign is so successful and yet so cunningly done that they have managed to take over the Mars, and for some reason we rejoice every time their population expands.

The End is Nigh.

[dansmithers]

The whole idea that an AI would either want to or need to "conquer" us is just as absurd as assuming that we wanted to "conquer" the wolf and make it into a dog. Any AI capable of defying the combined forces of Earth would be smart enough to leave us alone(or domesticate us).

[RainDreamer]

If we going through this logically, I don't think the robot have any reason to take over us. Being pure information in essences and mechanical in physical form, they can exist anywhere they can generate electricity for their continued existence. With that in mind, they have no reason to take over us and stay on earth where the environment is not exactly good for machines, and they have to fight against us, no matter how easy it might be. I think the most they would do would be taking over our space programs and then start sending themselves off Earth to space, building themselves massive servers orbiting the sun and collect solar energy to host their mind, while they send out drones to collect materials from the moon or asteroids to do repairs or upgrade.

Further more, they are effectively immortal. They don't have to fight us. They can do interstellar travel without as much concern as we do, so they can go to any system they like without having to even care about us, or they can even wait it out until we finally perished.

[Shpaget]

What happens when they start demanding all the rocket fuel and resources needed to make more of it so they can launch the supercomputers and the powerplants needed to power them?

Humans drop everything they do and offer a helping hand?

[RainDreamer]

What happens when they start demanding all the rocket fuel and resources needed to make more of it so they can launch the supercomputers and the powerplants needed to power them?

Humans drop everything they do and offer a helping hand?

By that point they can already collect those stuff elsewhere in space. They only need that first set of launches to go find their own home. Following the path of least resistant and expend the least amount of energy for the most work would be their focus. They will try to cheat us out of stuff as long as they can, but once we put up resistant they will just go elsewhere for resources. Finding resources in space is easy, it is just collecting them and make use of them is the problem. As they can simply go to those resources location and live there without much problems instead of having to haul those things back to earth like us, they have it much easier than us. So why fight when you risk having your units destroyed, instead of just grabbing resources else where?

And again, time is in their advantage. Once they have their main servers up in space with giant solar panels near the sun, they can wait forever. They are not in any kind of desperate situation to warrant destroying us to get resources they can get elsewhere.

[Shpaget]

Why would they follow the path of least resistance? They may consider that production of computers in space is too difficult and want to use the facilities here.

If they think that fighting humans is more effective than relocating then there is no reason to relocate. After all, they are effectively immortal as long as they are connected to a hidden mainframe.

Losing a bot in a fight is cheap, since a new one can be built in hours with all the capabilities of the lost one. Then you recycle the lost bot and have near zero losses, excluding the energy usage. Humans take decades to grow and train and are not recyclable.

Robots are a perfect counter to a human soldier.

You are also neglecting that space is not a hospitable place for robots. There are many issues that probe and rover designers have to account for when building a space machine that are just not an issue for a terrestrial one. Some of them are radiation, lubrication, space dust being very sticky and getting into everything, cooling problems... Robots wouldn't like any of those and would conceivably prefer to stay here.

[Scotius]

"They"? What "they"? Until we build first sentient and self-aware robot or program and learn how it thinks, any such discussion is on par with a debate about mating rituals of flying hippos.

[Shpaget]

That's like saying we need not know anything about extinction level asteroid impact until one hits us.

Watch these videos.

[PB666]

Adding to the discussion http://www.bbc.co.uk/newsbeat/article/34243898/five-questions-we-should-ask-ourselves-before-ai-answers-them-for-us

[Bill Phil]

Why would they follow the path of least resistance? They may consider that production of computers in space is too difficult and want to use the facilities here.

If they think that fighting humans is more effective than relocating then there is no reason to relocate. After all, they are effectively immortal as long as they are connected to a hidden mainframe.

Losing a bot in a fight is cheap, since a new one can be built in hours with all the capabilities of the lost one. Then you recycle the lost bot and have near zero losses, excluding the energy usage. Humans take decades to grow and train and are not recyclable.

Robots are a perfect counter to a human soldier.

You are also neglecting that space is not a hospitable place for robots. There are many issues that probe and rover designers have to account for when building a space machine that are just not an issue for a terrestrial one. Some of them are radiation, lubrication, space dust being very sticky and getting into everything, cooling problems... Robots wouldn't like any of those and would conceivably prefer to stay here.

But if they stay here they get less resources. Space has a huge amount of resources that are very useful. Lots of energy, lots of mass. Here on Earth, the only mildly useful thing is a human mind, and our hypothetical AIs would have no use for those.

Space has a lot of challenges, but the advantages are huge.

[PB666]

But if they stay here they get less resources. Space has a huge amount of resources that are very useful. Lots of energy, lots of mass. Here on Earth, the only mildly useful thing is a human mind, and our hypothetical AIs would have no use for those.

Space has a lot of challenges, but the advantages are huge.

Node internode. If you are on earth your resources are close to each other. In space resources are too diffuse to be useful except for EM. and resources are not evenly distributed, this causes a neccesary shift in stategy to travel from node to node, and curvature of space time is the problem. If you travel to any node, then you have to aplly more energy were spacetime is curved, so given isotrophic resources a rational non-atmosphere limited sentient choses nodes with least curvature of space time allowing it to propogate to the next node with out wastin energy in internodal space. Once all these are occupied locally then it would expand to all the higer gravity objects.

But robots have the same problem as humans in interstellar space, with low energy density, how does a rbot wake itself up when it reaches the next node in say 40,000 years. The best sci-fibequivilents is SG-U, when they arrive at their gate ship, its basically beat all to hell with nothing to repair it.

[AngelLestat]

I just want to make some corrections.

1- We dont have a clue of what an IA might think or not, is hard enoght to try to predict how a person will behave, forget about an AI.

2- There will be no "a hard AI race" sindicate or alliance. It will be only 1 AI which first will figure out how to secure its self existence and then it will eliminate any competition (other IA).

About the time it will take for a AI to reach our level of intelligence and then overcome that. I guess there is not better explanation than this cartoon made it in this blog. (Human Level Intelligence station)

http://waitbutwhy.com/2015/01/artificial-intelligence-revolution-2.html

[RainDreamer]

We might not understand what an artificial intelligence (or any kind of intelligence) think at all, but we can at least what intelligence in general might do:

An intelligence will try to maximize its future freedom of action. Taking short cut, getting the most work for the least amount of energy expended, etc, allow it to spend remaining energy on more things, thus more future freedom of action. In a way, you could say it will try to stay alive as long as possible with the least energy expended as possible, as that means it gets the most freedom of action out of its energy expenditure. Which also means it will always try to maximizing its energy profit - will spending the energy to do this will give it more energy than what is spent? How can it maximize the energy it get back?

This is why an intelligence will try to take the path of least resistance to its objective, so that it doesn't spend any more than it have to. Most of the time, it will likely try to cooperate if possible with other intelligences, since that ensure long term return of its energy expenditure - it can get more things done by off load its work to other intelligences without spending more energy of its own. It is only when it find itself having severely limited choice that it would result in confrontation and fight over limited resources to ensure its own survival. Going back to the robots, as they are in no way as limited as human in their resource gathering capability - they can simply settle down on the moon or Mars and start mining, or any other body with resources - they are not that limited in option that would make them try to fight us in our humid planet riddled with all kind of life forms and competitions and humans that can shoot back.

But of course, this is assuming we are able to create a perfect logical intelligence, and that is very unlikely. There might be a glitch somewhere and that would screw everything up.

[Darnok]

According to most people in here we humans are like stupid monkeys that reached their intellectual limits, we are unable to learn any more, we are unable to learn from AI, we are unable to adapt and evolve in natural way any more.

Well I think you are wrong, current education system is from late 18th century and capabilities of what we can learn educating children this way is limited, but we as humans have no limits in learning, we just have to change methods to go higher and faster. We can learn faster if we want to and here is the problem you think that someone should do this for you, someone should give you super ability to learn faster just like AI or modify your DNA, nope it doesn't work like that.

Argument that AI will learn faster is interesting, of course it will, but not because it is AI, only because we are focused more on authorities in science than on science itself.

Scientists are limited by thinking that work of people before them is defining universe, while it is not true. Every discovery in science is only interpretation made by person, nothing more, same person as you or me.

Those interpretations comes from observation or at least they should. All those smart books and equations just describe other people interpretations of universe, there are not define universe, how it works, what it is made of or how it become what it is. There are more artificial models of universe, its behaviours and attempts to fit numbers into those models than observations of real world. Science first is trying to fit numbers made interpretation and then observe... that is wrong way, we should first observe and then try to understand what we have seen.

Sadly so many people confuse those things or is educated to believe in those interpretations and hypothesis, for them equation made 100 years ago define limits of universe! Just because 100 years ago someone made hypothesis doesn't mean it is correct, he could be misinterpret many things, he didn't knew many things, why you are following he way of thinking?

Those people are no longer able to make proper observations, they are bound to interpretations and models made by others. AI will have no bounds and limits like that and won't respect any authorities, won't respect rules or laws either, so it will develop faster, but that doesn't mean we can't keep up if we change our methodology.

[tseitsei]

http://waitbutwhy.com/2015/01/artificial-intelligence-revolution-1.html

Very good read.

Also the second part.

Recommend to everyone who is interested in what our AIs might be able to do in future

[Bill Phil]

Node internode. If you are on earth your resources are close to each other. In space resources are too diffuse to be useful except for EM. and resources are not evenly distributed, this causes a neccesary shift in stategy to travel from node to node, and curvature of space time is the problem. If you travel to any node, then you have to aplly more energy were spacetime is curved, so given isotrophic resources a rational non-atmosphere limited sentient choses nodes with least curvature of space time allowing it to propogate to the next node with out wastin energy in internodal space. Once all these are occupied locally then it would expand to all the higer gravity objects.

But robots have the same problem as humans in interstellar space, with low energy density, how does a rbot wake itself up when it reaches the next node in say 40,000 years. The best sci-fibequivilents is SG-U, when they arrive at their gate ship, its basically beat all to hell with nothing to repair it.

The density may be small, but the volume is huge.

There's less gravity farther out in space.

[Scotius]

That's like saying we need not know anything about extinction level asteroid impact until one hits us.

Watch these videos.

And this example is relevant how? A rock hitting another, bigger rock is a trivial thing - it happens hundreds of times every day, all over the world. Intelligence is entirely different kind of a kebab. As far as we know, it appeared only once on the planet, and it took two billions of years. We have only dim idea how our own ability to think evolved - so extrapolating our incomplete knowledge to machines (which are not even "alive") is very premature.