Key points are not available for this paper at this time.
Artificial intelligence (AI) is reshaping catalysis research, yet its impact remains constrained by fragmented, heterogeneous, and non-standardized data. In this Commentary, we argue that the future of AI-driven catalysis depends critically on the development of data-centric infrastructures that integrate experimental results, theoretical insights, and literature knowledge. We highlight the Digital Catalysis Platform (DigCat) as an example of such an approach, combining large-scale curated datasets with tools for visualization, modeling, and machine learning. Building on this foundation, domain-specific AI agents enable automated data analysis, knowledge extraction, and catalyst design, transforming how researchers interact with catalytic data. We discuss how the convergence of big data, AI models, and automation may lead to closed-loop, autonomous discovery systems. Finally, we outline key challenges and opportunities in establishing reliable, interpretable, and scalable AI-driven workflows for catalysis in the data-rich era. Access DigCat at https://www.digcat.org.
Zhang et al. (Fri,) studied this question.