Manufacturers are under increasing pressure to improve resource efficiency and reduce waste in order to support circular manufacturing practices. Artificial intelligence (AI) has been widely adopted to facilitate this change; however, there is limited empirical evidence on the effectiveness of industrial digital tools. This paper presents a document-centric analysis of three representative industrial platforms, focusing on their data acquisition methods, AI-based functionalities, operational support and outcomes. The results demonstrate that these platforms address various stages of the manufacturing process. One category supports asset reliability through continuous monitoring and predictive maintenance; another enhances production line performance by reducing scrap and auxiliary resource consumption; and a third improves quality assurance through AI-enabled visual inspection, resulting in reduced rework, scrap, and emissions. The findings suggest that circular manufacturing is facilitated by the combined use of industrial connectivity, line-level analytics, and specialized AI applications rather than by a single, comprehensive solution. This study provides practical insights into a field of research that has predominantly been conceptual.