• • AI-driven virtual screening accelerates chiral catalyst discovery by over 10-fold, as evidenced by data-driven design of chiral carboxylic acids achieving >90% enantioselectivity in cobalt-catalyzed indole synthesis (Zhang et al., Nat Commun 2023).
• • Machine learning models predict chromatographic enantioseparation retention times with quantile geometry-enhanced graph neural networks, achieving median absolute error below 0.5 minutes, enabling rapid method development for chiral pharmaceuticals (Xu et al., Nat Commun 2023).
• • ML-guided synthesis of G-quartet-based circularly polarized luminescence materials achieved photoluminescence quantum yields up to 0.85, with dissymmetry factors (glum) reaching 0.02, outperforming conventional design by 40% (Dai et al., Adv Mater 2024).
• • Autonomous cloud labs with AI feedback control realized the discovery of optically active chiral perovskite nanocrystals, reducing experimental cycles from months to days and achieving circular dichroism anisotropy factors of 0.1 at visible wavelengths (Li et al., Nat Commun 2020).