Algorithmic mediation in the societies of control: characterization of user behavior modulation through machine learning techniques in Google patents

• Grant number: 23/01858-0
• Support Opportunities: Scholarships in Brazil - Master
• Start date: July 01, 2023
• End date: January 31, 2025
• Field of knowledge: Humanities - Sociology - Other specific Sociologies
• Principal Investigator: Pedro Peixoto Ferreira
• Grantee: Rafael Gonçalves
• Host Institution: Instituto de Filosofia e Ciências Humanas (IFCH). Universidade Estadual de Campinas (UNICAMP). Campinas , SP, Brazil

Abstract

This research project proposes an investigation on the characteristics of the processes of anticipation and modification of behavior through machine learning techniques on technologies patented by Google. At the end of 20th century, Gilles Deleuze proposed the term "societies of control" to name an emerging social logic that replaced the centrality of environments of enclosure and the disciplinary techniques applied to bodies, by flexible spaces and by mental and subjective control operations. In this context, there is a change of scope of capitalist system that begins to exercise power over, and explore economically, processes before their realization. This occurs through an alliance of capitalism with techno-scientific advances. One example of this is the patenting of genetic code, a potential organism. In a similar way, the emergence of artificial intelligence systems and associated machine learning techniques meant the creation of new control technologies capable of anticipating and affecting user behavior in situations mediated by algorithms. Therefore, we propose the analysis of eighteen patent documents from Google, the company considered the first and main agent of contemporary capitalism to develop and use such technologies. The purpose of the analysis is to characterize how machine learning techniques are used to anticipate and affect user behavior in technologies patented by Google.