## Neurophos focuses on optics for AI inference Neurophos is tackling the power efficiency problem in the AI industry with an optical chip that uses a composite material to perform the calculations required in AI inferencing tasks. The company has recently raised $110 million to advance this technology. ## Energy efficiency and inference Inference, the phase in which a trained AI model is used to make predictions or make decisions, is often a computationally intensive phase. Neurophos' optical chips promise to significantly reduce energy consumption compared to traditional electronic architectures. This could have a significant impact on data centers and edge devices, where energy efficiency is critical. ## The future of AI hardware Neurophos' approach represents one of several paths that companies are exploring to improve the performance and efficiency of hardware dedicated to artificial intelligence. Other areas of research include new materials, specialized architectures, and advanced cooling techniques. Competition in this sector is growing rapidly, with the goal of making AI more accessible and sustainable.