No abstract

One of the aims of this study is to present foreign, Czech, and Slovak projects created by generating literary texts through the use of artificial neural networks (we propose the term synthetic textual media). However, the main goal is to provide a critical analysis of the presentation strategies applied in the publication and promotion of the results of these projects. We will therefore test the hypothesis that these modes of presentation lead to the mythicisation of artificial intelligence and inappropriately skew the share of human and non-human involvement in the production of generative texts. We understand synthetic textual media and their presentational paratexts in complementary textual relations, and stress the necessity of critically analysing them as a whole. This stems from the fact that the current practice of literature generated by artificial neural networks is not suitable for a close reading approach without reductive reception and (mis)interpretation. We are also aware of the specificity of the reception processes initiated by literary texts of this kind and strive to support the concept of literary metareading, which we consider more appropriate for the technological and literary levels of this type of text.