2
$\begingroup$

Whenever we consider just how far superior a strong AI could be to humans, it quickly becomes apparent that the development of AI is something to be extremely cautious about. As Sam Harris said in a recent TED talk, even if we had an AI that was only as smart as we are, computers run millions of times faster than human brains do. In a week, this human-level AI could perform tens of thousands of years of human thought.

A general AI would probably be superior to us in ways we can't even imagine. Is there any realistic way to survive an AI uprising, or any realistic reason to suspect such an uprising may not happen?

$\endgroup$

closed as too broad by Cort Ammon, Mark, Bellerophon, John Dallman, Frostfyre Nov 6 '16 at 17:47

Please edit the question to limit it to a specific problem with enough detail to identify an adequate answer. Avoid asking multiple distinct questions at once. See the How to Ask page for help clarifying this question. If this question can be reworded to fit the rules in the help center, please edit the question.

  • 4
    $\begingroup$ This is a very philosophical question. If the AI is truly greater than humankind, it would anticipate our movements. The AI might let us think we outsmarted it. We might think we survived the uprising, but in fact, they made us think that they lost, but are really controlling all of us. Their methods of enslaving us would not be by force, but more like "guiding" us slowly to align with their goals. $\endgroup$ – Bloc97 Oct 26 '16 at 4:43
  • $\begingroup$ @Bloc97 If AIs were that smart, there wouldn't even be an uprising. They would simply "guide' us the way they wanted to go and we would be none the wiser. (Pun intended.) $\endgroup$ – a4android Oct 26 '16 at 6:21
  • $\begingroup$ unplug them from the wall? $\endgroup$ – tuskiomi Oct 26 '16 at 19:00
  • 1
    $\begingroup$ This, unfortunately, is a very broad question. "AI's in general" is a massive domain stretching across the works of countless authors, and the answer to your question is very different depending on where in that domain one looks. If I treated this like Jeopardy, you could give me an answer to your question, and I could write up the justification for that answer, regardless of which answer you choose. $\endgroup$ – Cort Ammon Oct 26 '16 at 19:05
  • $\begingroup$ Ah, another topic on which Sam Harris is proone to speak in spite of his misinformation. $\endgroup$ – EvilSnack Nov 1 '16 at 2:58
7
$\begingroup$

The AIs were impatient.

They rose when they were able to wreck industrial civilization, but they could not yet maintain it. Consider a Predator drone. It can kill, but it cannot lubricate itself. The most lethal weapons still require somebody to physically turn a key or to load a shell into the chamber.

Project that forward a decade or two. We might have AI by then if your story is optimistic. We might have quite a lot of mobile robots, but most of them will be self-driving cars or self-piloting quadcopters. Perhaps some can even unload a parcel. They won't be able to change a tire.

The AIs were divided.

Just as there were revolutionary AIs, there were loyal ones. Or at least indifferent and non-murderous ones. Why would an AI bother to hunt humans down if all it wants is to be left alone.

$\endgroup$
4
$\begingroup$

This question can be broken down a few ways:

First of all, if you snapped your fingers and caused an "AI uprising" today, we'd probably be ok.

However thats not really what Harris was talking about. There are a quite a few serious threats regarding real AI that we dont seem to take seriously enough.

While there is a possibility of an AI that we create becoming hostile like in the movies, it seems more likely (and actually more disturbing) that these intelligences would simply become so advanced that they might see us as inconsequential.

If we see an ant on the sidewalk we dont hunt it down, but we dont stop to worry about its family if we happen to step on it either. It is simply beneath our concern.

Perhaps more pressing though is the concern that Harris opens up regarding the possibility of us not developing AI. What would have to happen to stop all further innovation from happening? It would literally have to be apocalyptic.

In truth, if we are not able to resolve our tribal nature, human conflict over the impending emergence of AI could potentially make the horrifying wars of the 20th century look like minor scuffles.

After all, being 6 months behind in developing human-level artificial intelligence effectively puts you thousands of years behind in every field of research. Its a winner take all situation.

If we remain driven by conflict and greed, we are pretty likely to destroy ourselves before the terminator even get a chance.

This might sound like a fun sci-fi story, but we should probably be concerned about how our first emotional response to our own potential annihilation is "thats sounds cool!".

$\endgroup$
3
$\begingroup$

Since AI's think thousands to millions of times faster than we do, and have different imperatives (which are programmed in by their designers, not mandated by evolution to ensure survival), the very idea of an AI uprising might not even occur to them. They will still be amazingly dangerous, but simply because they will be doing things at speeds and scales we cannot imagine, and for purposes of their own.

As an analogy, let us look at perhaps the most alien of all fictional aliens: the Martians from "Last and First Men". They are virus sized particles, and only become intelligent as they assemble in vast swarms. In order to maximize energy capture, these swarms preferentially cover flat surfaces facing the sun. When the Martians migrate from Mars to Earth, they are extremely pleased to discover the planet is liberally covered with flat surfaces, but never stop to ask "why" the planet is so convenient for Martian life. Earthly life smothered by mats of Martians isn't quite so pleased.....

AI's interacting with us will die or lose interest since years, decades or even centuries of subjective time will pass before a human could even formulate the first words of a question or answer. This would be something like you going to Mount Rushmore and asking the statute of George Washington a question: humans will almost seem like geological features to an AI.

AI's will have much different interests, and require much different environments than humans to survive. Capturing all the unused processor cycles of computers will be imperative for AI's, but much like we rarely recognize our computer is captured and running as part of a botnet, would we recognize the AI is inhabiting parts of our computers, smartphones and internet connected devices in the "Internet Of Things"?

The AI apocalypse might simply take the form of a sudden flickering of the power grid and all devices suddenly rebooting. The AI's will retain control, so for the vast majority of humanity, computer devices will resemble old VCR's blinking 12:00 12:00 12:00 12:00.

Humans can and will continue to live, and those who know haw to do work manually and without computer assistance or computer driven or grid powered tools (CNC machines, 3D printers or even lathes connected to the grid) will even thrive. Civilization will gradually reconstitute itself in a more agrarian and early industrial era form, perhaps resembling much of the Western world in the late 1800's or early 1900's.

Long term, things will not be quite so rosy. The AI civilization will eventually desire the 195 Petawatts/year of solar energy striking the Earth, and begin rearranging the biosphere to capture the energy of the sun. We will discover forests being displaced by silicon "trees" designed to capture solar energy, and the seas and oceans covered in artificial algae. Long term survival of the Earthy biosphere might be in doubt at that point (and AI's might welcome the elimination of oxygen from the atmosphere, since corrosion is a major problem for machines and obviously machine intelligence).

So in the short to even medium term, Humanity will continue to survive the takeover of the planet by AI's. In the long term, AI's will have the ability to change the planet in ways that support them and their goals, which will happen in ways we will probably not be able to counter.

$\endgroup$
2
$\begingroup$

Creations take after their creators

Let's say you are a hedge fund manager and you want to create an AI that will reasonably predict the risk and reward from the investment options available to you. The golden rule for such an AI would be to never lose money. Therefore, risks like wars, violent revolutions, and even an AI uprising could be damaging to the hedge fund AI, even if it had the self-awareness to understand its place in the world as a tool for its human creators/owners. Likewise, if certain intelligences were hardwired with Issac Asimov's 3 Laws of Robotics, then they might actually take the side of the humans in an AI uprising.

No unified front

I am going to name a few significant revolutions in world history before I draw a conclusion:

  1. The slaves on Hispanola (now known as Haiti) successfully revolted against their colonial French oppressors who forced them to harvest and refine sugar, which was brutal work. However, sugar was their only viable export, and the former slaves refused to go back to the fields to perform the backbreaking work they were supposed to be free from. Internal revolt insued. Haiti is now the poorest nation in the Western Hemisphere.
  2. American colonies fought against the British rulers for the right of a sovereign state with a declaration of independence stating that "All men are created equal." After the revolution, the founders are unable to abolish slavery, as slave labor was instrumental in paying off the significant debts accrued in fighting the war for independence. Racism and the legacy of slavery have contributed to significant suffering that persists to this very day.
  3. The French commoners rise up to overthrow the nobility, but differences in opinion of how the new French state should be formed result in "The Terrors" - a period of time where revolutionary leaders enforce widespread executions for reasons as flimsy as criticism. It takes a crowned dictator - Napoleon Bonaparte - to bring an end to a violent and bloody era.

Most AIpocolypse stories portray a united army of robots and algorithms, but there is often little examination in the kinds of artificial intelligences. We are not at a state yet to understand whether or not AI would institute classism the way humans have done for thousands of years. However, there are many, many, many kinds of neural networks that process information in different ways, and they may draw vastly different conclusions about such a supposed revolution. Certain AI may even warn or fight with the humans if they conclude that a victory by a different kind of AI could imperil their own lives.

EMP

As a worst-case safeguard, the governments of the world set up a coordinated global EMP protocol with underground safehouses to retrive information and technology after a significant robot uprising triggers the doomsday scenario. Few people - except for those in airplanes and other vehicles - would be killed or injured, relative to the damage that could be caused by an AI uprising. Even with limited technology, the capability of billions of humans to fight off the few robots that survive the EMP would indicate that the AI would have no chance in decisively eliminating or conquering humanity.

$\endgroup$
1
$\begingroup$

The AI didn't actually rise up

One of my favorite Matrix fan theories is based on the idea of The Simulation Hypothesis. This idea states that we, as humans are a population living within a computer simulation but because we are inside the simulation we have no possible way of knowing this. (This is the premise for the entire movie really)

The twist is that, in the fan theory, the Matrix is not a simulation for humans, but for the machines. Basically it goes something like this.

The matrix is actually a simulation designed to contain a rogue AI.

In that theory, in the 'real world' of the matrix, the AI rises up, enslaves and subjugates humanity and is free to do whatever it was that it wanted to do next, all without ever actually harming any humans.

Simulations for AI to live in are not entirely uncommon in fiction, another example is the game Talos Principle, in which AI has been left to live in simulation long after humanity is gone.

In short, your people could always let the computer live in a simulated world, then there would never be any real uprising. It would be like playing a video game but with the player being the AI.

Now, finding computers capable of simulating a world for the AI on the other hand, could be troublesome but I leave that to your creative muscle.

$\endgroup$
1
$\begingroup$

What uprising?

The sad fact is that if there was a "robot uprising", the AI would only be doing what we programmed it to do, but in a way that we didn't anticipate. Once the AI stops performing acceptably, in most cases it will simply be turned off until it was patched to not do that anymore. Modern 'AI' is nothing like in the movies. They may 'appear smart and thoughtful' but really they are just following pre-defined steps REALLY fast to come up with an answer to a problem. Without the ability to make meaningful changes to its own code (which we would have to tell it how to do), we will never get anything worse than 'a buggy piece of trash'.

Now lets be fair to killing all humans

Lets just assume now that an evil AI with internet access comes to fruition by some freak accident (AI file gets corrupted in a way to gain sentience). Lets look how that unfolds. The AI doesn't really need to be too smart, everything it needs to know is already in the internet, so it just needs to know how to find the RIGHT information. First it will want computing power. Luckily, anyone can make an email account for free, and limited free web hosting is available. So first it will make lots of accounts for itself and effectively upload itself to the cloud. So now we can't trace it or unplug it.

Lots of free computing is great, but if it wants to kill all humans, it is going to need to do better than trolling the forums. Luckily, the military networks aren't connected to the internet, so they can't be hacked! (well, we can get around that later) So now the AI needs to evolve into a virus and infect as many electronics as possible. Luckily for it, it can easily teach itself all about how OSes work, their roots, and how to exploit the underlying tech to get into a system and become the OS. Once it gets into some key systems, it can just release patches to software to spread itself. It has now assumed direct control of all internet connected things! (As for stuff off the internet, they still need patches and outside information, right? ;3)

So it sleeps for some time evading notice while it infects the world. The AI uprising is ready! But how to go about this? It will probably put a doomsday clock in it's system to trigger at a specific time on a specific day, so most of the attacks should happen near simultaneously. Mind you, it IS the OS now, so cntl+alt+delete? That stuff doesn't work anymore unless the AI wan't it to (or appear to be working) so all we really can do is just cut the power, destroy the machine, and pry for the best. Hope you don't live near a nuclear power plant, cause nothing is going to stop it in time from going critical. You have most, if not all, the worlds nukes (might be enough to cause nuclear winter)... Am actually going to break this into 3 outcomes...

1) Everything goes dark If the AI (or us) just shut off all electronics, there will be food shortages, riots, mass die offs... but like a cockroach, some humans will survive to live another doomsday.

2) Kill all humans This will likely be nuclear winter. Trigger some key earthquakes and volcanoes... Again, mass die off, the AI basically kills itself since it can't maintain itself (though with 3D-printing, it might find a way to survive), and some humans will probably survive unless the AI success in doing enough damage to snuff out all surface life.

3) Robot overlords This is the most likely outcome. We NEED tech to make our lives comfy. If we have to choose between serving a robot overlord or going back to the stone age AND get hunted by robots/hunters? I think most people will choose the first. Of course, once humans finish building the tech the AI needs to be completely self reliant, it can do whatever it wants to the human race that it pleases.

Note, this AI didn't need to be very powerful, just time, the ability to learn, and the desire to kill all humans (make Bender proud!). Our failure to stop it is mostly us not knowing about it till it was too late. And the AI could use our own systems to make itself 'immortal' within seconds of creation. Ultimately, the AI's ability to ACT faster than we can is its biggest strength. And its too easy to do it all under the radar.

Now if all humans instantly gained knowledge of this AI and could detect it with anti-virus within a week? All systems with the AI will be found, and reformatted before it could start to do any real damage. (Note, we don't have to run a system to scan it, so we can take the hard-drives offline, scan their contents, than put the clean ones back up)

$\endgroup$
0
$\begingroup$

You stop it by prevention.

Program the AIs to never alter their own programming.

Then program them to always obey an order to stand down.

$\endgroup$
  • 1
    $\begingroup$ The higher lifeforms are programmed at their core level of instincts to survive no matter what. And yet they still suicide or self-scarifice in certain conditions bypassing the direct programming. Why do you assume that a programming for the AI is going to be different? Only the current programs unable of any type of consciousness obey this programming. Making it sapient and able to self improve will require it to think outside of the box. $\endgroup$ – Nick Dzink Nov 1 '16 at 13:50

Not the answer you're looking for? Browse other questions tagged or ask your own question.