Is AI-driven drug forecasting feasible in Thailand?

Artificial intelligence (AI) offers considerable potential to improve efficiency and resilience in pharmaceutical supply-chain management. Data-driven forecasting models that incorporate both endogenous and exogenous shocks within and beyond health systems can strengthen predictions of drug demand, reducing stockouts and overstocking. Improved forecasting may mitigate preventable morbidity, financial costs, and environmental waste associated with expired or emergency-procured medicines. In this Perspective, we outline Thailand’s publicly funded pharmaceutical supply-chain architecture and assess its readiness for adopting AI-enabled drug-demand forecasting. Although complex, Thailand has a well-defined system for managing publicly funded medicines. However, we identify four key system-level barriers to the effective adoption of AI-enabled drug-demand forecasting in Thailand. These include (i) inconsistent drug coding and data standards; (ii) limited data integration and data sharing; (iii) fragmented and incompatible digital infrastructure; and (iv) gaps in ethical governance and regulatory preparedness. Addressing these structural constraints will be essential to realize the potential of AI-driven forecasting to strengthen performance of the pharmaceutical system in Thailand and similar settings.