Brain-inspired phototransistor could cut AI energy use by sensing and storing data

Oregon State University researchers developed a brain-inspired phototransistor that integrates light sensing, memory, and signal processing into a single device. Published in Advanced Functional Materials, the hardware electronically controls the persistence or decay of optical memories. By combining these functions, the device aims to reduce data movement and improve energy efficiency for future AI vision systems. This innovation mimics human brain functions to optimize how digital memories strengthen or fade over time for better performance.