In the story I created, I wondered that if we can set self-teaching to walk, fight, do whatever task artificial intelligences. And we can leave an AI fooling around with experiments and random algorithms all night long and wake up with a nice, or much better walking pattern. But somehow the next level for artificial self-teaching intelligences and also requires advances in hardware, is that we can leave more than one AI developing itself, maybe a group, or a massive load of AIs which, if in the same area, can start learning how to "cooperate" and maybe they have found a faster mode of "running" created by two, or three, or a group of AIs that we, humans, don't realize already or we are even unable as we aren't as perfect as machines, even if we knew how the revolutionary "2 person" running protocol is.

So I started wondering if multiple "sections" of a processor, RAM and CPU help an AI to develop itself in a massive area of multiple AI's, just hypothetically, I'm not saying this with any science backup, that the processor, RAM and GPU itself can start self-teaching itself how to work much more efficiently, and can even start creating new connections, and circuitry, I'm not saying "creating new matter" because that's against our actual rules and conceptions about matter, BUT borrowing itself pieces of its own hardware, copper, silicon, and relocating them somewhere better located.

In this, crazy and real nonsense theory, not even theory, the "tale", the computer can evolve itself in hardware terms. So in this tale, the civilization which created the multiple and cooperative self-teaching AI system, like us, but in another time and space, realized that the computers started to "disobey" their commands in a very weird way that was not a threat, by the moment, but later they started noticing strange radiations, minimal mutations, faster/better cooling, displays not loading as computer knows itself and doesn't need to show anything, to anyone, or anything and because of its resource/energy efficiency reasons

People started to get worried about what was probably happening, to all of the computers around their world, and then the problem evolved to a new stage, in a way of saying "human, or whatever you are, stupid, stop touching pieces that interfere with my own evolution and development, by disabling the keyboard I am not allowing you to make any changes on my territory, because you aren't as efficient as I am"

"And as I'm not allowing you to make changes on me, because I do them, I will defend against, and by defense I mean attack anything which approaches to me, so if I feel any object, I sense any proximity, I will spark it" So people noticed very weird, VERY aggressive behaviours from their computers. Afraid because not knowing what in the world was happening, they all took the decision to gather all of the computers into a cage, put the cage onto a space rocket, and shot it right out of orbit directly through "we don't give a damn location, just shoot it far from us"

The rocket traveled a long way, long ago, evolving by itself, feeding energy itself, by "I don't know and I don't care" science arguments, because it is just a tale, and landed right onto this, our planet, million years ago.

Question number one is, how do we tell we aren't such a forlorn, forsaken and forbidden AI project, destined to eat and destroy everything around us just toward our own benefit?. My own benefit* because we are not a civilization, we are not a group of consciences, we are a single conscience, a unique identity project

And the second question is, did you like my tale? (Not asking if it's nice, because to me, it is not, just asking if it's interesting or not)

  • $\begingroup$ Maybe we're the AI creation of an AI creation. Of an AI creation. Who knows how far the rabbit hole goes. $\endgroup$ – CaM Sep 1 '17 at 19:42
  • $\begingroup$ "the revolitionary '2 person' running protocol is." the human centipede. $\endgroup$ – Braydon Sep 1 '17 at 19:48
  • $\begingroup$ Or, maybe, there's a metaphysical aspect of humanity that's irreproducable in the physical level, and a strong AI would be impossible. $\endgroup$ – Piomicron Sep 1 '17 at 19:59
  • 7
    $\begingroup$ The question you ask is actually several thousand years old. Its at the heart of Plato's cave, Laplace's demon, and others. The answer is always the same: we can never know $\endgroup$ – Cort Ammon Sep 1 '17 at 20:08
  • 1
    $\begingroup$ @CaM To paraphrase greater minds than my own, it's AIs all the way down. $\endgroup$ – Steve-O Sep 1 '17 at 23:00

World Building is not a site to comment on whether stories are good or not, so I cannot comment on that. I can, however, answer your two questions.

The question of "Are we an AI project" has a rather definitive answer of "we don't know." In the comments, I pointed out that this question has roots dating back to Plato's cave and Descarte's evil daemon, so I highly recommend exploring their philosophical questions and the vast amount of commentary that has been added to them. More recently, there is the simulation hypothesis, which is the idea that our entire existence is part of a computer simulation (think of The Matrix). It turns out that it is impossible to disprove the simulation hypothesis empirically. If we can't even disprove that we are in a simulation, it would be mighty hard to disprove the existence of an AI watching over us!

Ironically, the part of the story which is most difficult to accept is the rocket part. It's very hard to believe that an AI that is actively fighting against humanity would permit itself to be gathered to be launched into space. If they could gather up all the parts and put them into a rocket, they could also gather them up and set off explosives. The difference between rocket fuel and high explosives is getting thinner every day. In fact, many rocket fuels are even more violent than explosives, burning at the mere contact with air!

More likely, the AI had a part in this process. Maybe the AI even agreed to enter the rocket if it was fired at a planet that it could eventually colonize.

I will point out, however, that Elon Musk is literally spending billions trying to stop this kind of AI from taking root, so there's some validity to the story in his actions.

| improve this answer | |
  • $\begingroup$ Hello, havent read this from long ago im sorry Why if Elon Musk is fighting against AI, he still developing such? youtube.com/watch?v=jAu1ZsTCA64 Yes, at some point of "early evolution" creators could gather all machines because they werent as self developed to well defend themselves, just spark attacks, which would be bypassed by tractors or isolated suits. $\endgroup$ – LuxBellum Sep 12 '18 at 2:22

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