Why do people fall for conspiracy theories? Tokyo Institute of Technology Associate Professor Ryosuke Nishida Predicts “The Future of Humanity

People are just like that after all. In the first part, ” Hiroyuki’s Argument-breaking Art is a Symbol of the Times,” Ry osuke Nishida, 39, associate professor at the Tokyo Institute of Technology’s Graduate School of Liberal Arts, explains how to deal with the Internet and its mishmash of information. In the second part of the interview, we ask him about the usefulness of online discussions and the impact of “ChatGPT.

Until now, text information (written information) has been the main form of communication on the Internet, probably because communication was possible only through text, simply due to technical and service issues such as the size and speed of the connection. Young people are now using Instagram, TikTok, and short videos on YouTube as their main tools for communication. Technology and services are also moving toward photos and videos. In fact, even if you read textbooks on online and social network marketing, they tell you to use them to attract attention.

For better or worse, when people don’t read the same things, the cost of discussion becomes extremely high. To take one example from political science research, according to a study by Professor Akihisa Endo of Waseda University (Waseda University School of Social Sciences) and his colleagues, the definitions of “conservative” and “innovative” among young people are changing.

Conservatism” is associated with the ruling Liberal Democratic Party, and “innovation” with the opposition Communist Party, but when young people in particular hear the word “conservatism,” opposition parties come to mind. One reason seems to be that the image of opposition parties as people who oppose something in the Diet and do not change things = “conservative” has taken hold.

There is little point in saying that the perceptions of young people are “wrong”. It is not so strange that the way words are used change over time. If the perception of each word is different or the exact opposite, it will be difficult to establish a conventional discussion, and the cost of discussion should be high.

One of the problems with Twitter, which is transmitted in 140 characters, is that some people seem to think that the “short character count” leads to flame wars and dissension, but can a discussion be exhausted by producing long sentences?　I wonder. Some people post long sentences on Instagram or Facebook …… maybe they just don’t get read?

I think people get into conspiracy theories and obsess over things like bracketed “truths” simply because they are “interesting” or “eye-catching “. It is better to jump on data of dubious origin than to jump on news of relatively high validity reported by TV and newspaper reporters, because it gives us a sense of “proactive selection of information.

To begin with, disinformation and conspiracy theories may be tied to online business. Recently, it is also called hybrid warfare or cognitive warfare, and since it is sometimes related to the intervention of third countries, there are numerous (and numerous) ways to attract people’s attention.

–In recent years, the automatic generation of content, such as drawing using AI (Artificial Intelligence) and text generation using “ChatGPT,” has developed remarkably. As a result, many people are concerned that we will be further inundated with information of unknown origin and unidentified authenticity. What is the actual situation?

Since the cost of generating fake news is certainly lower with automatic generation, there is certainly a concern that such malicious information will increase. I think this is generally the consensus among media researchers.

When I was in the region on a speaking engagement the other day, we continued talking about ChatGPT at the izakaya where we launched the event, and the owner of the restaurant interrupted me, saying, “I use it too. He enthusiastically shared with me, in minute detail, his slightly wicked idea that I could use it when writing grant applications.

What I am trying to say is that it is amazing that a service that originated overseas has become so widely known among the Japanese population in such a short period of time. Even on TV, people are talking about “ChatGPT is amazing” day after day.

In terms of technological innovation, shifting to electric vehicles (EVs), for example, would certainly be more efficient, but there has been no progress. Japanese society is still dominated by the internal combustion engine. Why? Fuel efficiency may be good, but the initial cost of EVs is millions of yen, and they are more expensive than internal combustion engine cars, and modern internal combustion engine cars last quite a long time.

But ChatGPT and other generative AIs are different. The smartphones and browsers that run them are already widely available, many people can experience them immediately for free or at a reasonable price, and the impact is obvious. They have the potential to spread suddenly.

Many things may change in a society where generative AI is widespread: Bing, for example, will be able to pick up and deliver relatively valid information from the sea of information on the Internet, with links to NHK, newspapers, etc., when you throw in “tell me about the latest something”. It may not be the most necessary information, and there may be mistakes, but the number of times you come in contact with fake news or dubious information may be lower as a result.

The day may come when human beings may stop using their brains, but still be able to achieve high results: “smart people” who do not use AI or “not smart people” who use AI.

If the latter will be able to produce higher results, we will be blinded. Everyone will be flocking to the latter. When that day comes, education based on the traditional asymmetry of knowledge between teachers and students will be quite meaningless. What exactly should be taught at universities in such an era?

As an aside, what sometimes annoys me about generative AI is that it says, “There are N answers,” but the answers are reasonably correct but the number of bullet points is not N. (Laughs). I hope you can do something about that as soon as possible.

I am contrarian in my view that generative AI has the potential to advance human “cognition” and processing capabilities in a very fair way. Even in terms of the “anxiety” I mentioned earlier about not knowing what the right answer is, doesn’t it seem to you that we would be happier living in a world where generative AI is widespread, evolving, and sharing the correct information there and then widely and with many people?