Pedagogical conditions for understanding the text generated in the neural network

the authors attempt to develop an innovative approach in foundation and creating pedagogical conditions for understanding the text generated in a neural network. In this context, it implies systematic work with the generated text, where the received message (the generated text) is perceived as true or false, requires analysis and comparison with the original text (the original source). The article outlines the problems of understanding the text, its comprehension, interpretation and text generation. The difficulties associated with textual activity are described. The text is considered from the point of view of a generator of meanings, a mechanism generating meanings, a capacitor of cultural memory. The stage-by-stage work with the text includes not only the disclosure of meaning, but also the obligatory interpretation of what has been read. The considered theoretical aspects of linguistic, psychological and literary analysis of the text and theoretical approaches to understanding (meaning) of the text allowed us to identify the key pedagogical conditions necessary in the process of working with the generated text. The study was conducted from the point of view of the semiotic approach and the ideas of pedagogical hermeneutics. The research material was: M. Tsvetaeva poem «Our Kingdoms»; generated texts (according to the author's requests) in the YandexGPT 5 neural network. The paper presents one of the options for working with the generated text using the example of its analysis and comparison with a work of fiction in the educational process of secondary and high schools.

pedagogical conditions for understanding text, generated text, artificial intelligence, neural network