What's AI
AI is an abbreviation for artificial intelligence. It is a technology that allows computers to perform inferences, judgments, learning, etc. by imitating human intellectual abilities.
It has been almost two years since the release of OpenAI's AI service "ChatGPT," which answers questions as if it were having a conversation with a human.
Our society has been greatly shaken by the emergence of AI that can be easily used by ordinary people who are not IT engineers, and answers questions in the language we use at a search speed that humans can never catch up with.
The foundation of ChatGPT is an AI technology called generative AI. Conventional AI was researched with a focus on automating specific actions due to the limitations of computer performance. However, in today's world, where the evolution of machine power has made it possible to collect and process larger amounts of data, AI is now able to learn more than ever before.
The word generative, which is translated as generation, also means to create life. Generative AI can analyze trends from randomly collected data and literally generate completely new conclusions.
Let's take a concrete example. Previous AI could learn about cats by studying tens of thousands of images and literature related to cats, and classify, organize, and understand the concept of cats. However, generative AI takes a completely different approach.
By reading all kinds of data and learning the relationships and patterns in the data, it creates its own conclusion that a pet animal that has two eyes, walks on four legs, and has pointed ears at the top of its face is likely called a "cat." The more information it reads, the better it understands the information.
Even if it cannot distinguish between dogs and cats at first, the more it learns, the more it will understand the difference between dogs and cats. Therefore, generative AI can build its own theory of data classification and produce conclusions that humans would never imagine.
And those conclusions could lead to clues to solving environmental problems, energy problems, human rights problems, and other issues that humans have not been able to reach even after years of trying. While we are excited about the time-savings of office work, there is also this background behind the fact that large companies and countries are investing large amounts of money in researchers of generative AI.
However, since generative AI has the property of learning independently, there is a risk that it will evolve on its own out of our control. Mega clouds such as Microsoft's cloud service "AWS" provide a service that blocks information provision so that generative AI does not learn confidential information when a company uses generative AI.
This is now commonplace for mega cloud vendors. The fear of AI leaving human control is particularly strong in the West.
The background to this is the cultural background in which social norms have been created in accordance with the words of God written in the Book of Genesis in the Old Testament: "Be fruitful and multiply, and fill the earth and subdue it; have dominion over the fish of the sea, the birds of the air, and every living thing that creeps on the earth." Although it is not widely reported in Japan, demonstrations calling for an end to the development of generative AI are occurring all over the place, regardless of age.
