Towards AI/ML-Driven Optimization of Dendrimer Design, Synthesis, and Applications

Artificial intelligence and Machine learning (AI/ML) are of high interest for polymer chemistry and hybrid materials. The chemical composition, molecular, and macromolecular properties ultimately translate into their macroscopic properties and function. Various data sources, including scientific literature, are now augmented with large language models (LLM) via retrieval-augmented generation (RAG). These datasets serve as the basis for machine-readable descriptors, which can inform new molecular designs and synthesis routes. This perspective highlights directions and advances in dendrimer materials that can enhance data availability and the predictability of solutionand solid-state properties, as well as their applications in nanomaterials, energy, drug delivery, sensors, polymer processing and manufacturing. The theory and microstructure of dendrimers are revisited. AI/ML workflows, LLM and RAG applications, and the importance of the human-in-the-loop are emphasized. ML enables close computational