The global pharmaceutical drug delivery market is forecasted to grow to USD 2546.0 billion by 2029. The expanding pharmaceutical market urgently needs a more efficient drug research and development paradigm. Artificial intelligence (AI) is revolutionizing drug delivery by offering alternatives to traditional trial-and-error experimental approaches. This review systematically traces the technological evolution from early simple models to current advanced AI algorithms in various applications, ranging from formulation optimization to the prediction of critical formulation parameters and de novo material design. To enhance the reliability of AI applications in drug delivery, we present comprehensive guidelines and “Rule of Five” (Ro5) principles to systematically direct researchers in utilizing AI in formulation development. This “Ro5” includes the following criteria: a formulation dataset containing at least 500 entries, coverage of a minimum of 10 drugs and all significant excipients, appropriate molecular representations for both drugs and excipients, inclusion of all critical process parameters, and utilization of suitable algorithms and model interpretability. The review concludes with insights into emerging trends and future directions, including the utilization of large language models, multidisciplinary collaboration opportunities, talent development, and culture transformation, aimed at facilitating a paradigm shift toward AI-driven drug formulation development.

Drug delivery  /  Rational formulation design  /  Artificial intelligence  /  Machine learning  /  Deep learning  /  Formulation prediction  /  Rule of five  /  Multidisciplinary integration