When AI Discriminates: Comparing How PeopleRespond To Gender Discrimınation By AI And Humans

As artificial intelligence (AI) systems become increasingly integrated into differentdomains, their discriminatory behaviors become more relevant than ever. AIdiscrimination (i.e., AI’s discriminatory behaviors) can affect any group, but disadvantagedgroups, such as women, may be targeted more frequently. However,it remains unclear whether AI discrimination influences people’s attitudes towardstargeted groups and their willingness to challenge the status quo. Thus, this thesisexplores how AI discrimination affects people’s collective action intentions andintergroup attitudes. In two online studies, participants in Turkey (Study 1) andthe United States (Study 2) read about either (1) AI discrimination toward women,(2) humans’ discrimination toward women, or (3) an unrelated topic (i.e., controlcondition), and reported their perceived threat toward women, collective actionintentions for women’s equal treatment, attitudes toward feminists, and attitudestoward women. Findings revealed that AI systems were perceived as less threateningthan human discriminators but led to more perceived threat than the controlcondition. Contrary to expectations, neither AI nor human discriminator conditionssignificantly differed from the control condition in terms of collective action intentionsand attitudes toward women and feminists. These findings offer insights intothe group-level consequences of AI discrimination.