1 January 2026

Japan’s Long Return to Artificial Intelligence

Songruowen Ma and Kenddrick Chan

In the neon-lit laboratories of 1980s Japan, the future seemed already written. While Silicon Valley was in its infancy, Tokyo was already pouring billions into several major technology programs aimed at supporting cutting-edge scientific research. These included the Exploratory Research for Advanced Technology (ERATO), next-generation R&D program, and the Act for Strengthening Infrastructure for Research and Development of Industrial Technologies. Japanese companies took the cue, with Toyota starting to explore the auto-drive and voice command system, while the likes of Hitachi, Toshiba, and Panasonic had developed their own robotics divisions. The Japanese government also launched the “Fifth Generation Computer” project, aiming to create an “epoch-making computer” to surpass Europe and the United States in the field of information and technology.

Unsurprisingly, it was also in Japan that neocognitron, the architectural ancestor of the neural networks that power systems like ChatGPT, was developed by computer scientist Kunihiko Fukushima.

No comments: