Artificial Intelligence in Aquaculture: Current Applications and Future Directions

The aquaculture industry is rapidly expanding to meet the rising global demand for seafood. Traditional husbandry relies on manual handling and sampling, and heuristic rules that are labor-intensive, costly, slow, and often error-prone. As production scales, these methods are no longer practical or able to collect the massive amounts of data required to inform best practices. Artificial Intelligence (AI) offers a powerful alternative with the potential to enhance efficiency, accuracy, and scalability across the sector. Despite its promise, AI adoption in aquaculture faces key challenges, such as limited access to high-quality datasets and difficulties in integrating models with existing systems. To address these challenges and pave the way for more effective AI integration in aquaculture, it is essential to examine the current state of research, identify knowledge gaps, and highlight successful methodologies. This review introduces recent AI applications in aquaculture, with a focus on water quality management, feed optimisation, disease detection, genetic selection for breeding, and phenotypic analysis. Uniquely, this review (i) structures each domain around an end-to-end AI pipeline covering data acquisition, preprocessing, modelling, and deployment; (ii) compiles comparative tables detailing inputs, species, models, outputs, and performance; and (iii) synthesises cross-cutting challenges and future research directions drawn from the surveyed literature. The review highlights that deep learning architectures dominate predictive tasks, multimodal sensing is gaining momentum, yet clear guidance on reproducibility and edge deployment remains scarce.