Researchers Develop AI Semiconductor Using Hydrogen Migration for Learning and Memory
A team from DGIST has created an AI semiconductor that utilizes electrically controlled hydrogen migration for computation and data storage. This two-terminal device mimics biological synapses' learning and memory functions.
Theia Market Signal Identification - AI AssistedPublished Mar 16, 2026
Researchers from DGIST have developed the first AI semiconductor that employs electrically controlled hydrogen migration for computation and data storage. This innovation involves a two-terminal device designed to precisely inject and discharge hydrogen ions, thereby replicating the learning and memory functions of biological synapses. The findings were published in ACS Applied Materials & Interfaces.
Comments