The large language models (LLMs) represent a technology that has not yet seen widespread use. In addition to the automatic transcription of doctor-patient conversations, these tools will be able to support doctors in diagnosis and improve communication between doctors and patients. Their large-scale adoption is still limited due to concerns about sharing sensitive data, a lack of digital literacy, and the need to build new processes in clinical routines. To fully realize their potential, it is essential to train healthcare personnel on the functionalities of LLMs and to promote interdisciplinarity and scientific research, transforming artificial intelligence into augmented intelligence.