ROLE OF ARTIFICIAL INTELLIGENCE IN PHARMACOVIGILANCE: A CONCISE REVIEW

Abstract

The rapid expansion of drug safety data from clinical trials, spontaneous reporting systems, electronic health records, literature, and social media has placed significant strain on traditional pharmacovigilance systems. The analyzed articles collectively highlight the transformative role of artificial intelligence (AI) in addressing these challenges and enhancing drug safety monitoring. AI techniques such as machine learning, deep learning, and natural language processing have demonstrated substantial improvements in adverse drug reaction (ADR) detection, signal identification, case processing, and literature screening. Compared with conventional methods, AI-driven systems enable earlier signal detection, reduced false positives, faster case triage, and improved data accuracy. Several studies report notable reductions in case processing time and increased sensitivity in detecting safety signals, supporting more proactive pharmacovigilance practices. Despite these advantages, the articles also identify key limitations, including data quality issues, lack of model transparency, regulatory uncertainty, ethical concerns, and the need for skilled human oversight. Integration challenges related to infrastructure, validation, and global regulatory harmonization remain critical barriers to widespread adoption. Overall, the findings suggest that AI, when implemented within robust regulatory and ethical frameworks, can significantly strengthen pharmacovigilance systems. The convergence of human expertise and AI technologies holds strong potential to improve patient safety, optimize regulatory decision-making, and advance the future of drug safety surveillance.