Many high-stakes artificial intelligence (AI) applications target low-prevalence events, where apparent accuracy can conceal limited real-world value. Relevant AI models range from expert-defined rules and traditional machine learning to generative large language models (LLMs) constrained for classification. As the effort and expertise required to develop modern AI decrease, there is a risk that organizations devote too little time to understanding their limitations and sources of error. We outline key dimensions for critical appraisal of AI in rare-event recognition, including problem framing and test set design, prevalence-aware statistical evaluation, robustness assessment, and integration into human workflows. In addition, we propose an approach to structured case-level examination (SCLE), to complement statistical performance evaluation, and a set of considerations to guide procurement or development of AI models for rare-event recognition. We instantiate the framework in pharmacovigilance, drawing on three studies: rule-based retrieval of pregnancy-related reports, duplicate detection combining machine learning with probabilistic record linkage, and automated redaction of person names using an LLM. We highlight pitfalls specific to the rare-event setting including optimism from unrealistic class balance and lack of difficult positive controls in test sets-and show how cost-sensitive targets align model performance with operational value. While grounded in pharmacovigilance practice, the principles generalize to domains where positives are scarce, and error costs may be asymmetric.
Norén et al. (Wed,) studied this question.