A new AI chip that automatically adjusts its response characteristics to match the changing speed of incoming data has been developed in Korea, reducing prediction errors by up to 40 times compared with existing AI chips. According to the scientific community on the 8th, a research team led by Distinguished Professor Choi Shin-hyun of the School of Electrical Engineering and the Graduate School of Semiconductor Engineering at the Korea Advanced Institute of Science and Technology (KAIST) developed the "Programmable Dynamic Memtransistor (PDM)." Existing AI chips process data at a fixed response speed, limiting their ability to handle information changing at different speeds, such as rapid heartbeats or slow weather changes. The research team combined information storage with response speed adjustment into a single chip. Experimental results showed the PDM could adjust its response speed within a range of about five times, and in time-series data prediction, it reduced prediction errors by up to 40 times compared with conventional fixed-type AI chips. A chip integrating multiple PDMs maintained accuracy similar to existing software-based AI while operating with less power. The technology has high potential for mass production due to its non-volatile characteristic and compatibility with existing semiconductor processes (CMOS process). The research results were published in the July issue of the international journal Nature Communications.