Images of the (Non-)human Future in the Big Data World

Abstract

The variability of images of the future has expanded significantly with the advent of big data and unique generative pre-trained transformers that is a type of natural language processing models. The social imaginary gradually becomes sociotechnical, and agents of artificial sociality are included in the construction of various images of the future, reproducing stereotypes from the data used to train large language models. Big data and related technologies feature in various visions of the future, including utopias, dystopias, and realistic scenarios of an achievable and livable «good society». The emergence of the sociotechnical imaginary and generated images of the future initiates ethical and philosophical reflection, which is not limited to applied problems. However, there is also a need for exploratory studies, the results of which can be useful for further theoretical research. This article presents the results of a comparative analysis of the Russia future images generated by the GigaChat (Sberbank, Russia) and ChatGPT (OpenAI, USA). The utopia from GigaChat is focused on achieving technological leadership of Russia, and the utopia from ChatGPT is based on traditions. Dystopia according to GigaChat is a society of environmental disasters with a high level of socio-economic stratification and a low level of intellectual potential. ChatGPT constructs a Russian dystopia around an insidious small elite who, in various ways, including rewriting history, are trying to preserve their position of power. The «good society» from GigaChat is a utopia with restrictions, and from ChatGPT is an open multicultural society with a mixed economic system, focused on sustainable development and maintaining the mental health of citizens. The article demonstrates that different value and normative emphases in neural network images of the future are determined by the characteristics of the algorithms and the data used to train large language models.