Ayudando a mitigar riesgos en herramientas de IA generativa para Sistemas de Decisión Clínica (CDS)

A 2025 study examined the association between Sherpath AI usage and performance on HESI Custom Exams in an adult health nursing course at a public university in the southeastern United States. The analysis included 5,599 unique Sherpath AI queries submitted by 59 prelicensure students. Queries were classified using Bloom’s taxonomy and analyzed alongside students’ HESI scores across up to three exam attempts. Students who interacted more frequently with Sherpath AI demonstrated substantially larger gains on their HESI exam scores. High-use students improved an average of 400 points between their lowest and highest exam scores, compared with 194 points among low-use students.