Closed-Loop AI Enables Organic Continuous-Wave Laser

The realization of continuous-wave (CW) organic lasers critically depends on the rational design of high-performance gain media to mitigate thermal effects and optical losses under sustained excitation. However, such design remains challenging due to the lack of systematic frameworks and limited data availability. Here, we propose a closed-loop strategy that integrates theoretical modeling, deep learning, and generative artificial intelligence to overcome the limitations of traditional trial-and-error approaches. A geometry-aware generative model enables autonomous molecular design, while theoretical frameworks quantify the performance of gain media. This is coupled with state-of-the-art AI-enhanced screening to rapidly identify top-performing optical and electrical pumping candidates from a vast molecular space of 801,801 structures. Experimental validation confirms excellent gain properties, and most notably, continuous-wave laser emission is achieved for the first time in an organic thin film using a DFB resonator, with an ultralow excitation threshold of just 0.202 mW/cm² and stable lasing sustained for several hours. This breakthrough demonstrates the powerful potential of an AI-driven closed-loop workflow in scalable organic laser discovery and accelerates the development of high-performance organic solid-state laser technologies.