What question did this study set out to answer?

To assess the impact of ChatGPT on student performance during real-patient clinical exams.

April 23, 2026Medical Teacher0 citationsOpen Access

AI at the bedside: Randomised controlled trial of ChatGPT’s impact on student performance in real-patient clinical exams

HSHaroon Saloojee MGMichael C. Gramanie RMRhodasi Mwali

Puntos clave

To assess the impact of ChatGPT on student performance during real-patient clinical exams.
Conducted a randomised controlled trial
Evaluated minimally trained, self-directed use of ChatGPT at the bedside
Analyzed performance in real-patient clinical scenarios
ChatGPT use did not improve bedside performance
Any benefit requires structured training and prompts
Integration of LLM supports clinical competence but does not replace it

Resumen

Minimally trained, self‑directed point‑of‑care ChatGPT use did not improve bedside performance. Any benefit is likely to depend on structured training, consistent prompts or scaffolds, and clearer workflow integration. LLM integration can support, not substitute, foundational clinical competence.

Preguntar a la IA

Me gusta

Guardar

Ver artículo completo