Researchers have developed a foundation model called MOCLIP for designing nanophotonic devices. This model integrates metasurface geometry and spectra into a shared latent space using contrastive learning with a large dataset. The team demonstrated its capabilities in inverse design, generative optimization, and optical information storage, achieving high accuracy and density compared to commercial media. --- Why it matters: This matters to researchers in AI because it showcases the application of foundation models to nanophotonic design, which could lead to breakthroughs in fields like photonics and data storage. The model's ability to handle large datasets and achieve high accuracy is also relevant to the broader field of AI research. Source: https://arxiv.org/abs/2511.18980