Conceptual model of the use of a local AI assistant based on large language models to prepare responses to the public appeals to medical institutions

The paper is dedicated to the development of the conceptual model of the use of an assistant based on large language models to prepare responses to the public appeals to medical institutions. The key feature of the proposed approach is the intellectual system installing on a medical institution’s local server without transferring data to external services. This ensures compliance with the requirements of personal data protection and control of information flows. The article also discusses potential organizational effects, restrictions and risks of implementing this model in administrative practice of the medical institutions.