"Light-Powered Chip Revolutionizes AI Computing Speed"
Originally Published 1 year ago — by Phys.org
University of Pennsylvania engineers have developed a new silicon-photonic chip that uses light waves instead of electricity to perform complex mathematical computations essential for training AI, potentially revolutionizing computer processing speed and energy consumption. The chip's design combines nanoscale manipulation of materials with silicon, enabling it to perform vector-matrix multiplication at the speed of light. This breakthrough could lead to faster AI training and classification, reduced energy consumption, and enhanced privacy due to the elimination of the need for storing sensitive information in a computer's working memory.