Khushi Dahiya, Shikha Baghel Chauhan, Indu Singh, Chirag Jain
As a conceptual framework to address the well-documented pharmacokinetic and formulation problems of Cannabidiol (CBD), the integration of AI into CBD drug delivery research is being investigated. With a focus on theoretical potential rather than proven clinical validation, this paper critically analyzes new AI and machine learning-assisted approaches for the design and optimization of CBD-loaded delivery systems. We provide an overview of current understanding of CBD pharmacokinetics, which includes significant inter-individual variability, extensive first-pass metabolism, and poor aqueous solubility. We also discuss how formulation screening, dose prediction, and hypothesis generation can be supported by computational modeling and data-driven approaches. Preclinical research, simulation-based analyses, or early-stage methodological frameworks are the main sources of reported AI applications in pharmacokinetic modeling, patient stratification, and nanoparticle engineering; these should not be taken as proof of clinically proven efficacy. Although machine learning algorithms have demonstrated potential in identifying pertinent biological variables and optimizing formulation parameters, data quality, model interpretability, regulatory uncertainty, and a lack of prospective validation continue to limit their application in routine clinical CBD therapy. As a result, this review presents AI-driven CBD delivery systems as a developing field of study rather than a proven treatment option. The standardized datasets, model development, and experimental/clinical validations will likely be necessary in the future to determine if these computational approaches can significantly improve the efficacy, safety, and personalized application of CBD.