Defects, monitoring, and AI-enabled control in soft material additive manufacturing: a review

Additive manufacturing (AM) of soft materials is emerging as a disruptive technology for biomedical scaffolds, flexible electronics, and soft robotics, offering unprecedented freedom in design, compliance, and functional integration. However, three-dimensional (3D) printing of hydrogels, elastomers, and composite inks remains prone to multi-scale defects that compromise geometric fidelity, mechanical reliability, and biological performance. This review provides a comprehensive analysis of recent progress in understanding and mitigating these issues. We first examine the rheological and physicochemical properties of representative soft materials and their links to defect formation. Defects are then categorised into macroscopic, microscopic, and material-specific classes. State-of-the-art monitoring and detection techniques are critically assessed, spanning optical, thermal, acoustic, tomographic, and simulation-based approaches. In parallel, emerging data-driven strategies, including deep learning, diffusion models, and digital twins, are highlighted for enabling multimodal defect detection, predictive monitoring, and adaptive compensation. Persistent challenges in material transparency, multi-modal data fusion, and closed-loop intelligent control are discussed, alongside application-driven opportunities in personalised medicine, wearable electronics, and soft robotics. By bridging materials science, sensing technologies, and AI-enabled cyber-physical systems, this review outlines pathways toward defect-resistant, high-fidelity, and scalable soft material AM, and provides perspectives to inspire future research and industrial translation.