Analysis of the use of artificial intelligence language models in the automation of scientific communication

Abstract

The relevance of the study is determined by the rapid integration of AI language models into scientific communication processes, which is accompanied by a transformation of academic interaction formats and increasing requirements for the quality, reliability and integrity of research dissemination. Under conditions of intensification of automated text generation and knowledge exchange, the need arises to develop scientifically grounded approaches to the controlled use of AI in academic environments. The purpose of the article is to substantiate the theoretical and methodological foundations for the application of AI language models in the automation of scientific communication with an emphasis on ensuring the quality, reliability and integrity of research outputs. The research methods are based on theoretical generalization, analysis and synthesis of scientific sources, as well as a logical-analytical approach to interpreting the transformation of communicative processes under AI influence. As a result of the study, the conceptual essence and functional role of AI language models in scientific communication have been clarified, and the main directions of their application in the automation of academic interaction and knowledge exchange have been identified. Methodological approaches to evaluating the effectiveness and quality of AI-assisted communication have been substantiated, which include semantic, epistemic and communicative criteria. Key scientific and practical problems have been revealed, in particular semantic distortion, variability of outputs, transformation of authorship and insufficient institutional regulation. It has been proven that effective use of AI tools is ensured through their selective and controlled integration combined with mandatory verification of generated content. Conclusions indicate that AI language models function as instruments for optimizing scientific communication processes, provided that methodological control and academic integrity are maintained. Prospects for further research are associated with the development of standardized evaluation models, the identification of measurable indicators of quality and the empirical validation of the effectiveness of AI-assisted scientific communication.