The rational answer to the first question is: Humanity will perish. As a human being, we certainly do not want human beings to perish, but this is unavoidable by objective laws. There is an extended potential question for this question: Can people live forever? There have been countless common senses about this question since ancient times, and the answers obtained are all negative. Even with the current highly developed science and technology, many frontier propositions of scientific research seem to give people a little bit of hope, but it is logically impossible. It can be understood that on the one hand, scientific research also needs to attract attention, so as to attract investment. The subtext of immortality it expresses is most likely not for the real study of immortality, but for scientific research to obtain funding.
On the other hand, the media also needs an easy-to-understand, explosive news to attract readers and promote scientific ideas. This is very similar to the title of this article. There is a simple logical reasoning to show that the proposition “man lives forever” is a false proposition. Assuming that human beings can live forever as an individual, but the known earth and sun cannot be immortal, then for the immortality of the individual, it is necessary to ensure the immortality of the earth and the sun. If you leave the solar system and look for other galaxies, but is the universe immortal? Although a definite answer cannot be given within the known range, it can be seen that the energy required to ensure the eternal existence of an individual will increase with time. and a dramatic increase. is it possible?
The rational, personal answer to the second question is: very likely. In the process of learning artificial intelligence-related technologies, I have gone through several stages similar to a roller coaster in the expectation of artificial intelligence.
Stage 1: Curiosity. Curiosity is human instinct. I was very curious about artificial intelligence at first, and my mind was full of question marks about it. Kind of like looking at it diagonally.
Stage Two: Doubt. Based on the limited understanding and intuitive feeling of relevant computer knowledge, it is impossible for the programs written by individuals to achieve the capabilities described in the science and technology news! The word “frog at the bottom of the well” is the most appropriate to describe the current state.
The third stage: touch. This has to say about AlphaGo. This incident completely shattered all doubts in the second stage with facts. And this incident engraved an indelible sentence in my mind. The general meaning of this sentence is: The game that Li Shishi won created 2 firsts. He represents the first time that mankind has defeated artificial intelligence, and it is also the last time that mankind has defeated artificial intelligence. This is completely consistent with the core thinking of programmers: Talk is cheap, show me the code.
With the deepening of research, artificial intelligence has given complete solutions in CV (computer vision), NLP (natural language processing), RC (recommendation system), etc., and the combination of the above solutions and business has been incorporated to every aspect of life. Students can clearly feel the development of the Douyin Department. From the perspective of Zhuge Liang, the Douyin Department is based on uncertainty, and Ali and Penguin are based on certainty. Just like the development of information technology, it is developing from certainty to uncertainty (probability).
The fourth stage: On the basis of this trigger, the fourth stage is entered: hopeful learning. Begin to systematically study the theory of machine learning part at this stage, and put it into practice in some low-cost work life. In the process of this stage, I deeply realized that machine learning can actually solve the problems that could not be solved in the previous coding process, and it is a low-cost solution.
It needs to be emphasized here that the meaning of low cost is: someone has already carried the load for you, that is to say, the theoretical basis and practical operation part of it has been traversed for you, and what I am doing now is just the application. In the fourth stage, there are two outstanding experiences, which are directly expressed in two sentences: 1. The current machine learning is more like an art than a technology; 2. Before The knowledge of computer and communication learned is more like the moves in martial arts secrets; and after learning machine learning, I can see the way of generalized information technology (including computer technology, etc.).
In the process of this stage, I also deeply realized why the current era is called the biological age and the nuclear energy age? Is it the information age? The fundamental reason for my personal experience is that the mental method of information technology is mathematics. Information technology in a broad sense is built on a rigorous mathematical edifice. Generalized information technology is actually the realization of mathematical knowledge. This point can be clearly felt from the papers written by the three currently recognized computer ancestors-Shannon (does the old man on the cover have a taste?), Turing, and Von Neumann. List their three representative works “The mathematics theory of communication”, “Computing Machinery and Intelligence”, “First draft of a report on the EDVAC”.
If you have a little English foundation and academic background, you can clearly feel the domineering contained in the title (especially the first two papers). They don’t seem to be writing papers, but starting a discipline. Even standing at the moment, the profound thoughts of these three patriarchs still deeply shocked me. They seem to have traveled through time, seem to have seen with great certainty what they can do now, and seem to see further into the future. This is also a manifestation of the characteristics of mathematics: mathematics, it may not be guaranteed to give the most correct result, but it can certainly be guaranteed not to be wrong. In the process of learning, I also learned about the problems and applicable scenarios in current machine learning. The main manifestation is that its theoretical basis is not as beautiful as $F=ma$ and $E=mC^2$, but results are obtained through various constraints (manual repair). And especially the interpretability of deep learning is still one of the current research directions.
The fifth stage, in-depth study and research. Since we learned about the problems and applicable scenarios in current machine learning, it also pointed out the direction for future research. There is a saying: “If you are not afraid of thieves stealing, you are afraid that thieves will miss you”. The goal of artificial intelligence is to make the intelligence that humans are proud of can be achieved by artificial means. And now the goal of study and research is to contribute to this process.
In this stage of cognition, there is an understanding of human intelligence: human intelligence is a result of balancing various functions and performance parameters, it is not perfect, and its use is also subject to many restrictions. One of the greatest outstanding advantages of human intelligence is that it is very adaptable. That is, the generalization performance often said in deep learning. If the laws and characteristics of this adaptation can be found, then it is very likely that the next explosion point of artificial intelligence will appear. And is this the death knell of human extinction? However, from another perspective, this process is visible or controllable by human beings. Is this the luck of human beings? We have created a project in line with the Olympic spirit: more Is it a bit metaphysical to replace ourselves with higher, faster, stronger substances (species)?
To sum up, spacesrobot believes that the replacement of humans by artificial intelligence is a high probability event, and the only question is when it will happen. Let’s face it, you don’t know what the next piece of chocolate will taste like.
GIPHY App Key not set. Please check settings