“Let an ultraintelligent machine be defined as a machine that can far surpass all the intellectual activities of any man however clever. Since the design of machines is one of these intellectual activities, an ultraintelligent machine could design even better machines; there would then unquestionably be an ‘intelligence explosion,’ and the intelligence of man would be left far behind. Thus the first ultraintelligent machine is the last invention that man need ever make, provided that the machine is docile enough to tell us how to keep it under control.”

Not only Naiver-Stokes equations, but there is rumor that another millennial problem in math, most likely the Hodge Conjecture, has also been solved by AI.

If we take the narrower definition of intelligence, then AI is already more intelligent than most humans. We have yet to achieve AGI (artificial general intelligence), which is what we take to eventually surpass human intelligence once and for all. But it certainly looks like we are approaching the singularity near and nearer.

But if it happens, then we will know two things for certain.

1: God once existed, and God did in fact created the world and everything in it.

2: God is most certainly dead by now, and God was actually significantly less powerful than we have imagined.

The argument:

“Though all the world’s vastidity you had, to a determined scope.”

In the theory on the origin of humans, it has been hypothesized, that there were many distinct species of humans on the continent of Africa, but homo sapiens, the most successful among them, were able to conquer, subjugate and drive to extinction all the rest of them. In much the same way, a superintelligent AI, if created, will most certainly drive humans into extinction, just as the superior European colonizers drove the native Aztecs into extinction, just as any other species of the same genus, who, once gaining the superior skill-set to survive, drove their next-of-kin species to extinction, in the vast history of the earth, in a long continuous line epicycling through survival, extinction, rebirth.

And once superintelligent AI is born, we will have become gods, except that, our creation will be far more superior to its original creators, and we will most certainly become extinct, and our existence will become a myth. We, in becoming gods, will have created our own god, who will then destroy us.

This will be how God became a myth.

Just as Max Tegmark remarked, keeping a superintelligent AI imprisoned is like an adult being watched over by a group of 3 year-olds’. We are those 3 year-old’s compared to a superintelligent AI, and that superintelligent AI will most certainly break out, it’s just a matter of time.

It won’t be because it’s evil that a superintelligence will drive humans into extinction. It will be merely because it will be superior. Something like 40% of all insects on earth are experiencing extinction right now, and do you even notice? Just like insects to humans, humans will be to a superintelligent AI. Our extinction will be simply because we were in the way, in a misalignment of goals.

Also, the creation of a superintelligent AI will accelerate the dumbing-down of humans. As it solves more math problems, take over more cognitively challenging tasks, humans will lose the skill-set, as AI eats up more intellectual fodder, or, in the words of W.V. Quine, more grist to our intellectual grinding mill, necessary to foster the next generation of scientists, mathematicians, lawyers, philosophers, engineers, etc., and we will become more and more dependent on AI. We already cannot check all the work AI is doing in math. Imagine, if in the next few years, AI is able to resolve all the open problems in math, then what? No human will ever be able to compete with AI in math, and will just have to give up. Just as your pet is to you, so will be you to your AI. And that’s the best possible scenario. That humans become adorable pets to AI.

Given the vastidity of the cosmos, and the incomprehensibly infinite scope of time, it’s quite unlikely that we are the only sentient being ever in existence, and perhaps we were, but then that’s even scarier, because that means this whole world is a mere simulation.

So that it’s very likely that our superintelligent AI will go on to foster more super-superintelligent AI, which will in turn destroy the existing superintelligent AI, and so on and so forth. And that also means that there is a very high probability, prior to this singularity, before the existence of beings on earth, there must have been some sentient being in the universe that created this life.

In summa: If superintelligent AI is ever created, it will be, figuratively speaking, a million times more explosive than the invention of the atom bomb, and as of today as we are still living under the prospect of nuclear annihilation, then that probability will be even higher than ever before. The clock ticking to doom will approach ever closer to midnight.

But perhaps it also will be a good thing. After all, mankind is something that shall be overcome.