Gender Bias in AI: Creating Discriminatory Systems

Abstract

Through an examination of various societal factors influencing stereotypes, the chapter demonstrates how biases extend towards all gender identities. We acknowledge intersectionality by highlighting the connection between gender discrimination and other factors such as race, colour, and socioeconomic status. Solutions to address this bias in include improving diversity in AI development teams, increasing awareness, incorporating certain frameworks and policies. As such, addressing gender bias in AI needs a multifaceted approach, integrating ethical, societal, and technical considerations. By promoting diversity and a range of perspectives, developers can more effectively recognise and resolve biases present in AI systems. In order to create more reliable and morally sound AI systems, we also examine the significance of interdisciplinary cooperation between computer and social scientists, ethicists, and impacted communities. Finally, the chapter presents new research directions, which aims to create more ethical user system interactions.