Analyzing AI Bias: The Discourse of Terror and Sport Ahead of Paris 2024 Olympics

Abstract

Sporting events have been the target of terrorist groups for decades due to their high profile. Since 9/11, a bias against Muslims has been evident in both traditional and social media with Muslims being depicted as likely perpetrators of attacks against sporting events. The 2024 Paris Olympics represents the first major sporting events in the age of generative artificial intelligence (AI) platforms. This study is the first to examine to what extent AI chatbots replicate anti-Islam bias with respect to terrorism and sporting events. The study is based on a series of “conversations” with chatbots on the popular platform “character.ai.” The analysis reveals a persistent anti-Islam bias, with chatbots “explaining” that Islam is more prone to commit attacks because it is more violent in nature than other religions while also depicting Muslims as the most likely culprits of attacks against the Olympic Games. The study indicates that AI chatbots perpetuate the mediated anti-Islamist bias in an even more unanimous and authoritative tone than traditional and social media.