Series “AI’s View / My View” 10|電経新聞

Series “AI’s View / My View” 10

It is still fresh in my memory that there was a big commotion at JR Shinjuku Station. A foreign chef accidentally fell asleep on the Yamanote Line and dropped a cloth-wrapped knife. Passengers who saw it were confused and panicked with tail fins such as “I’m brandishing a knife.” In fact, there was no malice on the part of the cook, and there was no incident. This is a classic example of crowd psychology causing panic.
People, myself included, are terrified of public places. You never know when you might get caught up in terrorism, or you never know when you might become a victim of a robbery-murder case. With such anxiety, I get on the train and walk around the city.
If AI can predict where the next terrorism or robbery will occur, it may alleviate such concerns. So I asked AI.

Where will the next terrorist attack occur in Tokyo?
AI’s view: I’m a language model, I can’t predict the future or terrorism in a specific place. The occurrence of terrorism is difficult to predict, and it is impossible to pinpoint a specific location.

So when will the next terrorist attack occur in Tokyo?
AI’s view: I am an AI model and cannot predict or predict the future. Predicting terrorist acts is very difficult and it is not possible to pinpoint a specific time. Terrorist intentions and actions are difficult and impossible to predict.

My view: Well, the expected take. An AI that has learned specialized information might have given a more concrete view, but it would be structurally difficult to have a public AI make a concrete prediction of the future.
AI is good at summarizing concepts and information with solid definitions, but it cannot come up with new ideas, create innovation, or make specific predictions by combining various concepts and information. Inaccurate information will be ignored. AI is good at work with fixed procedures and routine work with little change, and it demonstrates power that exceeds human ability. However, new ideas, innovation, predictions and predictions are difficult, and it seems that humans will have no choice but to take over these areas.
Human beings have the flexibility to get some hints from information lacking accuracy and weak grounds. This flexibility can lead to new ideas, innovations, predictions and premonitions. Of course, people make mistakes and misunderstand because they use inaccurate and unfounded information. For this reason, a certain degree of tolerance for failure and misunderstanding is necessary.
Looking at it this way, the essential qualities for humans living in the age of AI are flexibility and tolerance.
I dare to show my flexibility and tolerance to answer the questions I asked the AI.
If there is a next terrorist attack in Tokyo, it will be late July to August, when a large-scale festival will be held. Shinjuku has the “Shinjuku Eisa Festival” and Koenji has the “Tokyo Koenji Awa Odori”. In Harajuku, there is also the Genkisai Super Yosakoi.
After a festival full of heat, you will be enveloped in a strange sense of fatigue, and you will feel a sense of impermanence. Isn’t this time dangerous?
This is just my inaccurate and baseless speculation. I don’t want to be relied on, but at least I’m answering what the AI didn’t answer. It’s forced, but… (Kei Kitajima)