Answers are grounded in this paper and cite their sources.
Artificial intelligence (AI) is playing an increasingly central role in drug discovery and the pharmaceutical industry more broadly. However, despite some high-profile success stories, many technically successful pilots do not translate into sustained business value. In this study, we analyze the determinants of successful AI adoption in the life sciences through three complementary approaches: (i) brainstorming workshops with senior industry professionals, (ii) a global executive survey, and (iii) a statistical analysis of a collection of AI/ML use cases collected by the Pistoia Alliance. Across all methods, a consistent pattern emerges: high strategic importance and demand for AI contrast with low organizational maturity, resulting in limited success rates for production deployments. Statistical analysis reveals that business success correlates strongly with project maturity, business support, and developer support, i.e., dimensions of organizational capability maturity, while factors such as data sources or model types are not predictive on their own. The findings reinforce that AI projects follow the same success drivers as other complex engineering initiatives and that organizational readiness, change management, and alignment between stakeholders are critical. Furthermore, trustworthiness encompassing reproducibility, explainability, and governance, is identified as a limiting factor for adoption, particularly in high-risk domains. We synthesize these observations into a set of recommended practices that emphasize fitness-for-purpose method selection, integration of AI into human workflows, and development of shared standards.
No takes yet. Share an insight, caveat, or question.
Makarov et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: