AI-driven robotics for optics

Optics experiments are essential across science and technology, yet they remain predominantly manual, limiting throughput, reproducibility, and scalability. Automating such experiments is challenging due to strict precision requirements and the diversity of setups in typical real-world optical laboratories. Here we introduce a platform that integrates generative artificial intelligence, computer vision, and robotics to automate free-space optics experiments. The system translates user-defined goals into valid optical configurations, assembles them with sub-millimeter accuracy, and performs micron-scale fine alignment using a robotic tool. It then executes a range of accurate measurements, including beam characterization, polarization mapping, and spectroscopy, with consistency surpassing human operators. This work establishes the first flexible and generalizable framework for optics automation, enabling programmable and adaptive experimental workflows.