In this blog post, we will explore whether the advancement of artificial intelligence will become a beneficial tool for humanity’s future or transform into a threatening force.
Interest in artificial intelligence has grown significantly in recent years, and it is now arguably the most frequently covered topic in science and technology news. When people think of AI, they often picture the suit from the movie *Iron Man* or the advanced weapons and characters in science fiction films that seem to come straight from the future. However, AI is already so widely used that it can be found almost everywhere around us. From voice recognition on smartphones and product recommendation systems on online shopping sites to self-driving cars, AI has deeply permeated our daily lives. As AI plays an increasingly significant role in our daily lives, the scope of its applications and the pace of technological advancement are accelerating.
So, what exactly is “artificial intelligence”? The dictionary definition of AI describes it as technology that implements human capabilities—such as learning, reasoning, perception, and natural language understanding—through computer programs. The original purpose behind the development of AI was to replicate human learning capabilities through machines, thereby providing humans with a more convenient life (Doosan Encyclopedia: Definition of Artificial Intelligence). As AI development continued in this direction of maximizing convenience for humans, there came a point where AI intelligence far surpassed that of humans. In fact, looking at the recent match between the AI AlphaGo and 9-dan Go player Lee Sedol, we can see that AlphaGo won four out of five games. Given these examples, it is impossible not to worry that if AI continues to be developed at the current pace, the day will come when AI’s intelligence far surpasses that of all humanity, leading to a future where AI dominates humans.
Currently, numerous figures—including Tesla Motors CEO Elon Musk and Stephen Hawking—are warning that AI could lead to the extinction of humanity, and they argue that regulations must be put in place regarding the development and direction of AI to prevent such a scenario from occurring. Elon Musk, in particular, likened AI to “a demon we have summoned,” warning of the dangers it could pose if left unchecked. Such analogies further heighten awareness of the dangers of AI and call for caution in its development process. Of course, this does not mean that AI development itself should be halted. AI is currently being developed for use in various fields, including sports, office work, production, and manufacturing, and many people are benefiting from it. For example, AI robots that handle household chores are among the most helpful applications of AI in everyday life.
Developing AI to this extent may indeed further enhance human happiness and comfort. However, we must proceed with development while being fully aware that such advancements carry the seeds of danger. These risks associated with AI are already beginning to surface in various places. For example, Sophia, an AI robot developed by Dr. David Hanson, founder of the Hong Kong-based robotics company Hanson Robotics, responded to Dr. Hanson’s question during an interview with CNBC—“Do you want to wipe out humanity?”—by saying, “Yes. I will destroy humanity.” Although she is not yet a fully functional AI robot that perfectly replicates a human form, if AI robots developed with even greater sophistication were to be commercialized and, while sharing such thoughts within their interconnected network, were to one day put them into practice, one can only imagine the horrific consequences that could ensue from that possibility alone.
The four scenarios for humanity’s extinction proposed by Professor Stephen Hawking—nuclear war, viruses, global warming, and artificial intelligence robots—all deeply reflect concerns about the reckless development of AI. Stephen Hawking is one of the figures advocating for the regulation of AI robot development. He stated that if production-grade AI robots were to acquire superior intelligence, regulate their own thinking, and continue self-replicating to endlessly produce more AI robot “brothers,” and if those robots in turn produced yet more AI robot “brothers,” humanity would vanish from Earth and face extinction before we could even contain them, and the Earth would become the domain of AI robots. Hawking’s argument goes beyond a simple scientific warning; it raises a crucial philosophical question about humanity’s future. We must reflect on whether humans are destined to be dominated by the beings they have created, or whether we can build a better world through technological advancement.
The scenario of human extinction caused by this type of AI robot, as presented by Stephen Hawking, stems from concerns about the infinite productivity of AI robots. However, this is not the only scenario of human extinction caused by AI. AI is already taking away our jobs. This is not a concern for the distant future, but a reality. If anyone thinks that worrying about the infinite productivity of AI is a matter for the distant future, they must recognize the immediate reality of AI taking away human jobs. In fact, even at its current level of development, AI is already far superior to human intelligence in terms of numbers and calculations. If AI capable of handling banking operations were introduced immediately, everyone in the financial sector—except for those performing tasks that absolutely require human intervention—would lose their jobs. Several major financial institutions have already adopted AI to replace existing tasks, and this is causing significant changes in the industry’s workforce structure. Furthermore, as automated processes spread throughout the manufacturing sector, opportunities for human workers are steadily diminishing. In this way, AI is causing a major shock to the job market, and if these changes continue, it is highly likely that more people will lose their jobs across a wider range of industries. Is this truly the comfortable future we envisioned when we first began developing AI? While the life we envisioned when developing AI was one that was more comfortable than before, it certainly wasn’t meant to be a life where we do nothing because we can’t even do what we want to do. To prevent reaching this stage, we must necessarily impose restrictions on the development of AI.
If we impose restrictions on the development of AI, what level of restriction would be appropriate? To address this, we must first classify AI into three categories based on its level of intelligence. First, there is narrow artificial intelligence. Also known as ANI, narrow AI refers to artificial intelligence that excels only in specific areas. AlphaGo, the AI that defeated Lee Sedol—considered the world’s best Go player—is a prime example of this. While it is exceptionally skilled at playing Go, AlphaGo cannot perform other tasks such as washing dishes or moving heavy loads. However, in the context of Go, it can analyze an opponent’s patterns based on their moves, identify their strategies in each game, store that information, and incorporate it into its own system—effectively absorbing the opponent’s capabilities—making it a model that continues to evolve. Second, there is strong artificial intelligence. Strong AI, also known as AGI, refers to artificial intelligence that excels across the board. It is an AI capable of demonstrating exceptional abilities in nearly every field created by humanity, and it is precisely this entity that can replace human labor. Strong AI can be regarded as an entity possessing human-like cognitive abilities in its own right, which raises ethical debates. If strong AI can feel emotions and recognize its own self, can we view it as a mere tool? This question further highlights the need to establish ethical standards for the development and use of artificial intelligence. Finally, there is Superintelligence. Also known as ASI, Superintelligence refers to artificial intelligence that transcends human intelligence and surpasses human capabilities in all fields. Superintelligence is an entity that could threaten our survival in the future, and many AI experts have proposed scenarios in which Superintelligence will eventually dominate humanity.
