Chinese Universities Develop 3D Photonic Neural Network in Glass for AI Computing
Researchers from Huazhong University of Science and Technology and Shanghai Jiao Tong University have developed a programmable photonic neural network embedded in glass, achieving a theoretical throughput of 6554 TOPS. This innovation, published in Nature Communications, marks a significant advancement in optical computing by utilizing light not just for data transmission but as a direct participant in computation. The chip's architecture involves a three-dimensional photonic lantern waveguide array and a programmable phase shifter array, enabling complex optical processing of two-dimensional images with high accuracy. 
