Precision meets intelligence in AI-driven LPBF: transforming design, flexibility, quality, and sustainability in additive manufacturing

The integration of laser powder bed fusion (LPBF) processing with data-driven design, process control, and quality assurance, mostly mediated by emerging artificial intelligence (AI)- based toolkits, is reshaping the landscape of additive manufacturing (AM). LPBF offers advantages such as design flexibility, rapid prototyping, cost efficiency, and high-resolution processing compared with mainstream AM methods. Despite early success owing to its unrivalled geometric precision and microstructural control, the LPBF lacks industrial scalability, intrinsic process control and real-time monitoring, resulting in unpredictable defects, reduced efficiency, and poor overall process control. There is a clear opportunity to overcome many of these issues by integrating an AI-based software layer into the LPBF process, creating a powerful hybrid software-hardware system that combines the best of both worlds. The hybridization of smart software of AI with precision engineering of LPBF signifies the arrival of a paradigm change toward sustainable production rather than just an incremental technical advancement by eliminating excessive material waste, lowering energy usage, and increasing product dependability when compared to traditional approaches. This review carefully curates studies that specifically targets the data-driven protocols, for holistic closed-loop integration of AI-enabled LPBF across the entire manufacturing ecosystem. Special emphasis has been placed on design for additive manufacturing (DfAM), laser parameter optimisation, in-situ process monitoring, closed-loop feedback control, defect prediction, and sustainability assessment. The specific objectives of this review are: how AI helps interpret sensor data to better monitor the printing process; adaptive laser control for optimised material structure, properties, and performance; and the potential to revolutionise quality control, not only by identifying issues but also by progressive learning to improve future prints and thereby steadily outperforming traditional methods. In addition, emerging technologies like as generative artificial intelligence, digital twins, and miniaturised laser systems are discussed as potential enablers of autonomous and environmentally responsible production. In summary, this article establishes AI-assisted LPBF as a strong, intelligent, and future-ready manufacturing paradigm.