Learning from AI Winter: The History of AI and Its Discontents in South Korea

After seeing how the fierce enthusiasm for developing Artificial Intelligence suddenly

turned into sarcastic disillusionment in the late 1970s, a group of Korean scholars who had

studied AI abroad - especially in the United States - came back to South Korea in the early

1980s with a cautious attitude toward praising the power of AI. They organized a study group

at the Korean Institute of Information Scientists and Engineers in 1985, and published a

periodical newsletter to introduce accurate and precise information about the development of

AI without giving any groundless illusion. Their newsletters became the first and central

place in which many experts in Korea participated in articulating a particular form of

discourse and practices on AI. This paper shows how the issue of AI had been addressed and

framed in Korea, being partially affected by the experience of AI winter in other countries,

through analyzing a series of AI newsletters published in the late 1980s. In addition, this

paper reveals how this cautious attitude on AI began to be changed with the rise of neural

network theory in Korea. It was the neural network theory, which emphasizes the importance

of understanding the human brain to develop AI that had regarded as a promising and

favorable research item for Korean scholars. This paper explores the way in which AI had

been discussed in Korea which later influenced on developing and utilizing the expert system,

neural chip, and fuzzy system in the 1990s.