AI for Materials Discovery

Artificial Intelligence (AI) is revolutionizing materials discovery by accelerating the design, synthesis, and characterization of novel materials with targeted properties. By leveraging machine learning models trained on experimental and computational datasets, AI enables predictive design, structure-property correlation mapping, and autonomous experimentation. Recent advancements, including generative models, graph neural networks, and large language models, have enhanced our ability to explore complex chemical spaces efficiently. Integrated with high-throughput synthesis and automated characterization, AI-driven workflows can potentially enable the discovery of functional materials for energy, electronics, and sustainability applications. I will illustrate our foray into using AI in the whole materials discovery process, including atomistic simulations, property predictions, generative design and experimental optimization. I will also provide an overview of where the field stands and why this remains an unsolved problem.