A new study published in Nature reveals that researchers have successfully demonstrated a significant improvement in the energy efficiency of a key AI computing process. The team developed a novel chip architecture that performs vector-matrix multiplication, a fundamental operation for neural networks, using photonic components instead of traditional electronic transistors. This approach reportedly reduces the energy consumption of these calculations by over a hundredfold compared to current state-of-the-art electronic chips. While still a laboratory prototype, the technology points to a potential path for overcoming the growing energy demands of large-scale AI systems. The full research details are available in the article: https://example.com/full-article