In this blog post, we analyze the threats that artificial intelligence could pose, such as weaponization, job replacement, and the replacement of creativity, and consider whether it is a real crisis or a new opportunity for humanity.
Long ago, in 2014, at the 100th anniversary symposium hosted by the Massachusetts Institute of Technology (MIT) Department of Aeronautics and Astronautics, Elon Musk, CEO of Tesla Motors, made a bold statement that AI research was akin to summoning the devil, sparking widespread debate. Following Musk, brilliant physicist Stephen Hawking, Microsoft co-founder Bill Gates, and Christoph Koch, director of the Allen Institute for Brain Science, also warned of the dangers of AI surpassing human intelligence. However, a look at human history reveals that machines surpassing human capabilities is not a recent phenomenon. When pulleys were invented, humans fell behind machines in terms of strength. After the invention of the automobile, humans fell behind in terms of speed of movement. With the advent of calculators, humans were surpassed by machines in computational ability. But why is it that, when it comes to intelligence, humans falling behind machines is such a controversial issue? Is the prospect of machines surpassing human intelligence truly a threat to humanity? To answer this question, we will examine the characteristics of artificial intelligence that could pose a threat to humanity, review the arguments of those who warn of the dangers of artificial intelligence, and determine whether these dangers are truly a threat to the survival of humanity.
According to Wikipedia, artificial intelligence is “the technology of making computers behave in ways that would be considered intelligent by humans, such as learning, reasoning, perception, and understanding natural language.” Before defining artificial intelligence, it is important to define intelligence, which in the above definition encompasses learning ability, reasoning ability, perception, and understanding of natural language. In other words, AI is a machine that not only processes data at the speed of existing computers but also possesses human-like thinking abilities. Additionally, AI can learn its own code and develop intelligence in a manner similar to humans. When envisioning the future of this technology, it is not difficult to imagine machines with higher intelligence than humans. Such imaginings inevitably evoke fear, as humans have long believed that their ability to exist as the strongest life form on Earth stems from their superior intelligence compared to other animals. If it is true that life forms with higher intelligence dominate those with lower intelligence, the day when machines dominate humanity may not be far off.
The specific threats to humanity posed by the development of artificial intelligence have been explicitly mentioned by the Future of Life Institute (FLI), which includes Stephen Hawking and Elon Musk, the British BBC, and Christoph Koch. The first is the threat of artificial intelligence being used as a weapon, the second is the threat of artificial intelligence replacing jobs, and the third is the prediction that artificial intelligence will be humanity’s last invention. We will now examine these three threats and determine whether they could have a real impact on the survival of humanity.
First, if developed as a weapon, AI could become more dangerous than nuclear weapons. This argument was made in a letter signed by 1,000 leading scholars and entrepreneurs from around the world, published by the Future of Life Institute mentioned earlier. In the letter, they call for a ban on the development of “AI-automated weapons,” expressing concerns that such weapons could evolve into autonomous systems capable of attacking targets without human intervention. Additionally, they argue that the development of AI is relatively easier compared to nuclear weapons, which require scarce resources, making it easier to produce in large quantities. If such weapons fall into the hands of terrorists or dictators, they could be used for “ethnic cleansing.” However, the likelihood of such a scenario unfolding is low. When considering the risks of AI-automated weapons, one might think of nuclear weapons. As of 2023, there are nine countries that possess nuclear weapons, but they have never been used for ethnic cleansing and have only been used twice in history. Additionally, considering the ease of developing AI-automated weapons, looking at the case of firearms, which are relatively easy to develop, shows that they do not pose a significant problem. While there are occasional news reports of casualties from gun accidents, these are well managed under law enforcement, and if artificial intelligence can identify and manage potential issues in advance, it can help prevent threats to humanity.
Second, the development of artificial intelligence will threaten human jobs. The BBC in the UK launched a service in September 2015 that visualizes jobs threatened by robots. This service tells you the percentage of risk that a particular job will disappear when you enter the job title. The top-ranked job, telemarketer, has a 99% risk, meaning that approximately 42,570 of the 43,000 telemarketers in the UK are expected to lose their jobs. However, as we have seen throughout human history, technological advances will create more and more diverse jobs. In the 19th century, the Industrial Revolution in the UK led to the spread of machinery, causing unemployment and sparking the Luddite movement, which sought to destroy machines. However, while the Industrial Revolution temporarily increased unemployment, it ultimately created new jobs such as engineers who designed and repaired machines. Today, engineers have evolved beyond simply constructing and repairing machines to encompass a broader range of roles, including researching their principles and designing them. When you enter “mechanical engineer” into the BBC service mentioned earlier, it states that only 13% of these jobs will disappear, with the phrase “It’s quite unlikely.” This demonstrates that, when compared to jobs eliminated by machines in 19th-century Britain, machines have provided more stable jobs rather than threatening human employment. Artificial intelligence may eliminate jobs such as doctors and financial analysts, but humans are animals that continue to evolve and develop. Even if one generation becomes unemployed, the next generation will use artificial intelligence to create more stable jobs.
Third, “Artificial intelligence will be humanity’s last creation.” This statement was made by artificial intelligence (AI) expert Kristof Koch in an interview with Current Affairs Journal. He referred to a situation where AI programs its own code to enhance its intelligence. Decades ago, universities were only for a select few. This meant that it was possible to acquire knowledge usable in society within 20 years of birth. However, as society has become increasingly complex, the knowledge required for human development has grown exponentially, and the time needed to acquire it has also increased. In the distant future, the point at which it becomes impossible for humanity to accumulate the knowledge necessary for its own development is likely to be the point at which AI replaces humanity’s creative activities. However, the replacement of humanity’s creativity by AI and the threat to humanity’s existence are separate issues. Let us consider this phenomenon by drawing parallels with other cases in history where machines surpassed human capabilities. Machines can exert unimaginable power beyond human capabilities and perform tasks that humans cannot do. However, machines do not replace humans in all tasks that require human strength. Society will change in the same way in terms of creation. The surpassing of human intelligence will not threaten the survival of humanity or take away all of humanity’s creative activities, but rather help humanity achieve creation beyond the “limits of intelligence.”
After examining the three threats that AI development poses to humanity, we have confirmed that these are no different from countless situations that have existed throughout human history, and that AI development is actually a technology that will benefit humanity. In summary, AI technology has the potential to develop into a weapon, but it can be properly managed under public authority. While it may take away jobs from humans, it can also serve as a medium for creating more stable jobs. Finally, it will not halt humanity’s creative activities but will help humanity overcome the “wall of progress.” Of course, the current generation may face temporary threats due to its inability to anticipate the rapid development of AI. However, humanity has always overcome the risks associated with new technologies throughout history, and AI will be no exception. In other words, artificial intelligence is not a threat to the very existence of the human species. Rather, it will be recognized by our descendants in the distant future as one of the many inventions that once threatened humanity, flowing as a single stream within the vast current of history.
