Based on Yuval Noah Harari’s Sapiens, this course explores the possibility of AI replacing human intelligence. We’ll analyze the debate over whether A.I. can have the same understanding and ethical judgment as humans, and look ahead to the future of human intelligence.
In Yuval Noah Harari’s book Sapiens, he warns that the end of Homo sapiens is near, and one of the reasons is that the development of cyborg technology will make us no longer human. In fact, the transformation from human to cyborg is currently underway, thanks to the scientific revolution. Even now, we can replace body parts with machines, such as prosthetic hands and legs, and as bionic technology becomes more advanced, we will be able to create replacements that perform as well as living organisms. For example, when an organ fails, instead of removing tissue or transplanting someone else’s organ, we may soon be able to use artificially created replacements. However, there is still a lot of debate about whether AI can replace human intelligence.
AI has become an indispensable part of modern society. Whether it’s the AlphaGo match between Lee Sedol and AlphaGo a few years ago, the self-driving cars that are currently the talk of the town, or the appliances that use the word “AI” in their advertising, artificial intelligence is closely intertwined with our daily lives. If this rapidly advancing technology can achieve human-like intelligence and problem solving, it could one day replace humans. However, there are some people who deny the possibility of AI reaching human intelligence. They argue that AI cannot replace human intelligence for two main reasons. First, it may be able to do the same things as humans, but it won’t know what it’s doing or what it means. Second, they argue that AI cannot satisfactorily address questions of value or ethics. However, both of these arguments are refutable, and it’s entirely possible that AI will become equal to human intelligence.
The first argument that AI cannot replace human intelligence is that, unlike humans, AI does not know meaning. Humans can derive laws from phenomena, infer new facts, and even create new things by applying given rules. On the other hand, those who argue that AI cannot equal human intelligence believe that AI is just a complex piece of software with an output determined by its inputs. This argument is based on John Searle’s “Chinese Room” experiment. In this experiment, a non-Chinese speaker in a room can follow a set of questions and answers and have a normal conversation with a Chinese person outside, but the person doesn’t understand Chinese. This could be interpreted to mean that while an AI can produce accurate output for an input, it does not “understand” the rules in between, and cannot use those rules to create new outputs.
However, the “Chinese Room” experiment may be a false analogy. Using the human brain as an example, individual neurons do not understand language, but the human brain as a collection of neurons does. Similarly, even if the people in the room don’t understand Chinese, the room as a whole as a system does. This proves that AI can have the same understanding as humans, and that it can be equal to human intelligence. Furthermore, by developing algorithms for finding problems or deriving laws, it will be possible for AI to understand the meaning of the problems it solves and have the intention to solve additional problems that arise. This would refute the claim that AI can only do one thing.
A second argument that AI cannot replace human intelligence is that it cannot adequately address issues of value and ethics. They argue that because AI doesn’t make value judgments like humans do, if it were to replace human intelligence, many problems would arise. Since many of the world’s problems are related to values and ethics, and AI cannot solve problems without prioritizing certain situations, it is impossible for AI to replace human intelligence. There are simply too many cases to define every situation.
However, AI is now at a stage where it can learn on its own. Our sense of ethics and ability to make value judgments is not a product of evolution, but rather the result of learning from a young age and experiencing different situations. AI can learn just like humans, and because it is a machine, it can experience more situations. Therefore, the potential for AI to solve problems involving values and ethics is very high. It is also possible that A.I. could become superior to human intelligence because it can make decisions in a very short time, whereas humans are inconsistent in their value judgments due to indecision and deliberation.
The argument that AI will never equal human intelligence may have been convincing in the past, but it’s less so today. Current AI learns by mimicking human brain structures (neurons), such as perceptrons, and draws conclusions. Claims that AI doesn’t understand what it’s doing, or that it can never equal human intelligence, contain logical fallacies or stem from a lack of understanding of modern technology and strong AI. In this essay, we argue that AI has as much understanding as humans and can reach the same level of human intelligence. Therefore, it is quite possible that AI can replace human intelligence. However, there is still a lot to think about when it comes to replacing human intelligence. Even if humans become less human, AI should be utilized after ethical and institutional developments have been made. As Yuval Noah Harari wrote in Sapiens, we must ask the question of what we want to become.
