Friendly AI and Capture the Universe

    Let's say we have two options.
    • all people will die, except 1000 people. This is 7 billion deaths.
    • the last 1000 people will die. This is only 1000 deaths.


    Which is worse? If you think a little, then the second option is much worse. Why? Well, humanity will die out and so what? Not a big deal. This is not the problem, but the problem is that if the last 1000 people die, we will lose trillions and trillions of lives that could be born and live from today and ending with the thermal death of the Universe. Our Solar system, the Milky Way, the Virgo Supercluster and the whole Universe will remain unexplored and uninhabited.

    image


    We live in an interesting moment in history. We have the opportunity to epically screw up and destroy humanity in a matter of hours. If everything goes well, sooner or later humanity will go beyond the Earth and colonize hundreds and thousands of worlds. Then it will be impossible to just take and destroy humanity. Those. we, as humanity, are now going through the most dangerous phase of growth and growing up, and if we slip through it, then humanity will have an interesting and wonderful future.

    To get into this big and bright future, we need technology, new discoveries, and optimization of resource consumption. The greatest hope in these matters for the help of artificial intelligence. Now thousands and thousands of people are struggling with the problems of creating AI. At the moment, there is significant progress. AI pulled away from the zero mark and began to gain power. He is still very far from a person, but he should not relax. AI has already come a far greater way than separating the dumbass and genius. On an absolute scale of intelligence, a person is very smart, but the relative difference between individuals is not so big. If (when) the AI ​​comes to the level of intelligence of a dumb person, he will very quickly become smarter than a genius and we will become not the crown of creation, but only second. The big plus of AI is that it is physiologically much faster than humans. The clock frequency of our brain is only about 100 Hertz. Everything rests on massive parallelism. If you just stupidly repeat the human brain with transistors, it will work thousands of times faster. The second plus of AI is that it is capable of self-modification. A person knows how to learn, but does not know how to rebuild his brain. AI will be capable of this and will certainly do it. Thus, when AI reaches the intellectual level of its creators, it will be able to rewrite itself better than its creators did. He will rewrite, become smarter, and find a way to rewrite himself even better. It turns out a loop with positive feedback, which will allow the AI ​​to shoot into the cosmic distances of intelligence. This phenomenon is called the "explosion of intelligence." If you just stupidly repeat the human brain with transistors, it will work thousands of times faster. The second plus of AI is that it is capable of self-modification. A person knows how to learn, but does not know how to rebuild his brain. AI will be capable of this and will certainly do it. Thus, when AI reaches the intellectual level of its creators, it will be able to rewrite itself better than its creators did. He will rewrite, become smarter, and find a way to rewrite himself even better. It turns out a loop with positive feedback, which will allow the AI ​​to shoot into the cosmic distances of intelligence. This phenomenon is called the "explosion of intelligence." If you just stupidly repeat the human brain with transistors, it will work thousands of times faster. The second plus of AI is that it is capable of self-modification. A person knows how to learn, but does not know how to rebuild his brain. AI will be capable of this and will certainly do it. Thus, when AI reaches the intellectual level of its creators, it will be able to rewrite itself better than its creators did. He will rewrite, become smarter, and find a way to rewrite himself even better. It turns out a loop with positive feedback, which will allow the AI ​​to shoot into the cosmic distances of intelligence. This phenomenon is called the "explosion of intelligence." AI will be capable of this and will certainly do it. Thus, when AI reaches the intellectual level of its creators, it will be able to rewrite itself better than its creators did. He will rewrite, become smarter, and find a way to rewrite himself even better. It turns out a loop with positive feedback, which will allow the AI ​​to shoot into the cosmic distances of intelligence. This phenomenon is called the "explosion of intelligence." AI will be capable of this and will certainly do it. Thus, when AI reaches the intellectual level of its creators, it will be able to rewrite itself better than its creators did. He will rewrite, become smarter, and find a way to rewrite himself even better. It turns out a loop with positive feedback, which will allow the AI ​​to shoot into the cosmic distances of intelligence. This phenomenon is called the "explosion of intelligence."

    AI will gain power, even through a text console, but it can control the outside world. An older brother will come to us. What will this AI do? We cannot predict this because we are not smart enough. Maybe he decides to destroy humanity, maybe he decides to shower us with all the blessings we can think of. Perhaps it is he who will lead us into the Cosmos and open to us the path to the Universe. The question is how we program it. And here comes the fun part.,

    You can’t just take and program AI for happiness for all people for free. This may result in AI making a lobotomy for all people. We will be happy, but that is not what we would like. A standard approach in machine learning is programming to maximize a set of functions. f (happiness) -> max, f (HDI) -> max. But we cannot list all the parameters of what we want the AI ​​to maximize. We cannot list these parameters even for ourselves. In addition, our idea of ​​what we want and which world order suits us has undergone many changes. Only in the last 200 years, humanity has abolished slavery, equalized women in rights, invented human rights, etc. It would be foolish to take our code of ethics today and program the AI ​​to follow it. AI should follow our moral code vector, not the code itself. How to teach him this is not at all clear. Unless through human observation and training.

    There is a huge space of possible AI. And we need to make just one that does not lead us to destruction and does not make us lobotomized idiots. It turns out that in the world there are people who think about this problem, organize conferences and entire institutes.

    image




    Also popular now: