Researchers are exploring the use of large language models in clinical diagnostics, with a focus on grounding these models in real-world clinical data to improve their accuracy and explainability. Studies are evaluating the performance of large language models in various medical applications, including pathogen surveillance, epilepsy diagnosis, and laboratory test interpretation. The effectiveness of large language models in clinical diagnostics is still being assessed, with some findings suggesting they may not yet match the accuracy of traditional decision support tools.