COMPREHENSION THE PERPETUATION OF HISTORICAL RACIAL BIASES IN SOCIAL MEDIA ALGORITHMS ALONGSIDE CHANGES IN DIGITAL PLATFORM GUIDELINES

• Grant number: 25/06320-4
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: July 01, 2025
• End date: June 30, 2026
• Field of knowledge: Applied Social Sciences - Communications - Journalism and Publishing
• Principal Investigator: Angela Maria Grossi
• Grantee: Kauã Bernardo Coelho
• Host Institution: Faculdade de Arquitetura, Artes, Comunicação e Design (FAAC). Universidade Estadual Paulista (UNESP). Campus de Bauru. Bauru , SP, Brazil

Abstract

Algorithmic racism refers to the reproduction of racial prejudices in automated decision-making systems, be they recommendation algorithms, facial recognition or content moderation. These prejudices often emerge from the interaction between biased databases and the development of algorithms that do not take into account the complexities of racial and cultural identities. In turn, digital colonialism denounces the concentration of power and technological resources in the hands of Silicon Valley companies such as Meta and X (formerly Twitter), which design business models and cultural norms that are imposed globally, often to the detriment of local diversity and the digital sovereignty of other regions of the world. The central theme of the research is the impact of changes in the content moderation policies of these platforms, with an emphasis on the relaxation of moderation under the argument of "freedom of expression", as justified by CEOs Mark Zuckerberg and Elon Musk. Trumpism, as a political phenomenon, is configured as a macro meaning in the context of big tech, influencing and making explicit the legal and ideological loopholes used to allow and even foster hate speech, especially racist, on social networks. It aims to understand how the rise of Trumpism and changes in the stance of technological giants allow the perpetuation of forms of algorithmic oppression and the proliferation of hate speech in a digital society marked by an unstable balance between freedom and responsibility. The methodology will be exploratory, with a qualitative bias, by means of a bibliographical and documentary review and a case study of the Meta and X platforms, in order to verify their usage policies and the presence of biases that could constitute algorithmic racism. This research aims to broaden the discussion on racial biases and the role of digital platforms in content moderation.