Artificial Intelligence for Fluorite Ferroelectric Materials: From Discovery to Optimization

Artificial intelligence (AI) has emerged as a fruitful tool in materials science, enabling accelerated discovery, characterization, and optimization of functional materials. Among ferroelectrics, fluorite-doped hafnium oxide attracts exceptional attention due to its CMOS compatibility, scalability, and robust ferroelectricity at a few nanometers in thickness. However, understanding and optimizing the relationships between metastable ferroelectric phase formation and property-processing remain challenging due to the multidimensional parameter space governing its synthesis and performance. This review examines how AI methodologies, ranging from machine learning-assisted first-principles simulations to deep-learning analysis of experimental data, are reshaping the study of HfO<inf>2</inf>-based ferroelectrics. While such approaches have advanced understanding of structure-property relationships, AI-driven synthesis process optimization, and closed-loop synthesis remain underexplored. We outline current achievements, identify critical gaps, and propose next steps that integrate multimodal data fusion, active learning, and combinatorial synthesis to enable predictive design and autonomous optimization of next-generation hafnia-based ferroelectrics.