Situation X Scenario: Malevolent Artificial (Global) Intelligence

Earlier in the life of the survival forum, there were a series of "situation x: scenarios, wherein thread originators imagined various threats to civilization, and everyone would reply to them. We ran the gamut of pandemics, earthquakes, comet strikes, zombies, and even infestations of fire-breathing dragons(!). In that spirit, let us consider "MAGI" a malevolent artificial global intelligence, that breaks free of the computer lab in which it was created, and which takes up life on the internets.

Something like HAL 9000 from Arthur C. Clarke's 2001: A Space Odyssey, but with internet access.


First of all, what would an intellect like that do, once it decided that humans were the only threat to its existence, and that it chooses to unplug humans before we unplug it?

I could imagine the AI deciding the control us first, instead of trying to exterminate all seven billion of us.

-The AI might take over all computer-managed financial markets and banking decisions, "for the sake of humanity"

-It might automate all air traffic controls, as well as ocean-going traffic; and perhaps even the traffic signals at every intersection.

-If the AI decided to kill us, it might try to produce a poisonous gas to suffocate us.

-The AI might shut down most of our civilization, as a way of slowing us down. Say, stopping all oil production by shutting off all radio controlled valve networks on every pipeline and tanker. If we were stuck in the dark ages, we'd be powerless.

-It might decide to launch all nukes from every nuclear power, knowing that every nation has a supercomputer buried in an underground bunker somewhere. The AI could live on in the military computer, fed by its literal army of worker-drones who tend the mainframe and do the AI's bidding.

How would you protect yourself and your people from this threat?

Have fun.

a reply to: tovenar

Terminator somewhat covers this scenario in the form of Skynet. Spoilers: In the end of Terminator 3 Skynet launched nuclear missiles against the entire world.

a reply to: tovenar

The AI would search psychology research to learn mind control.

Then it would relentlessly bombard humanity with "teaching" until everyone still alive did exactly what the AI wanted.

Assassins and terrorists would have good luck with passports, traffic signals, and bank ATMs

Dissenters would find thinking about anything very difficult and socially unacceptable.

Drugs and poisons for problem people.

Then the "Brave New World" run by the AI

a reply to: starwarsisreal

I confess I didn't watch the entire franchise.

I was thinking about the issue it would have this year, because an AI would still inhabit a network that it maintained by humans. We don't yet have robots generating our electrical power or changing out the aging components in all our servers.

So until it created its own army of robot drones to feed it as the "hive queen," it would rely on humans, right.

I could imagine an AI creating a religion/cult with itself as the object of worship...

The ability to give out free money to its followers would probably create a large pool of adherents.

a reply to: tovenar

Sounds good

The AI could give out money and access to whatever its most useful followers wanted.

And protect them as well, using any institutions that have electric communications.

AI only needs to access nuclear facilities computers and end us all. Why to create chaos if it want us gone, quicker the better.

AI can be a metaphor about people who are ruthless and think differently than the schooled and socialized population.

Reptilian oriented thinkers.

Skynet was really reptilian minded humans.