Attributes of generated text in academic discourse: the problem of identification

Abstract

The current stage of academic discourse development is characterized by significant transformations associated with the introduction of artificial intelligence (AI). The face of academic discourse is changing, which requires analyzing and identifying relevant socio-communicative activities within the discourse. A need has emerged to critically conceptualize human-AI interaction within academic interaction, namely in terms of expert identification of generated text based on linguistic features. The aim of the study is to identify the linguistic features of the generated Russian-language text in academic discourse for the identification of oral and written academic texts. The scientific novelty of the study lies in the fact that for the first time on the basis of bibliometric indicators a comprehensive list of signs of the generated Russian-language academic text, obtained as a result of analyzing domestic theoretical and experimental studies (author's experiment with a neural network with a given prompt) is presented. The results of the study allow us to demonstrate the linguistic markers of Russian-language generated text within the framework of academic discourse at a given synchronic linguistic slice.