Transparency in Artificial Intelligence for Key Opinion Leader Identification in the Pharmaceutical Industry: Advantages, Limitations, Lessons Learned, and Emerging Regulatory Frameworks

The identification of Key Opinion Leaders (KOLs)—clinical experts who influence treatment guidelines, evidence dissemination, and therapeutic adoption—remains essential for pharmaceutical engagement and medical affairs strategies. Artificial intelligence (AI)-based platforms, including Eularis, Axtria, and IQVIA, facilitate efficient processing of multimodal big data sources such as scientific publications, clinical trials, social networks, insurance claims, and real-world evidence, enabling dynamic, predictive, and scalable KOL profiling. These tools promise to overcome limitations of traditional manual methods, yet concerns persist regarding algorithmic opacity, potential biases, variable data quality, and regulatory compliance in frameworks such as GDPR, HIPAA, and COFEPRIS. This perspective systematically reviews the advantages of these platforms (e.g., enhanced efficiency and scalability in data-rich contexts), alongside critical limitations (e.g., lack of explainability and risks of erroneous or biased recommendations). Historical case studies, including the commercial failure of IBM Watson Health and methodological challenges in the UK COVID-19 Infection Survey involving IQVIA, highlight the consequences of insufficient transparency. Recent regulatory developments, including the 2026 joint FDA-EMA guiding principles for good AI practices in medicines development, are discussed as foundational for future accountability. Recommendations emphasize adoption of explainable AI (XAI) approaches, internal audits of data sources and algorithmic weights, continuous human oversight, and alignment with emerging global standards. Achieving balance between technological innovation and ethical responsibility is imperative to ensure AI serves as a reliable complement to human expertise in healthcare decision-making processes.