Black artificial intelligence chatbots were perceived as more competent and more human by consumers than White or Asian AI bots during a marketing simulation, according to a new study that suggests traditional racial stereotypes that exist in real life may play out differently in the digital environment.
The study was led in part by Vanessa Perry, professor of marketing and strategic management & public policy, and Nils Olsen, assistant professor of organizational sciences, at the George Washington University. Participants were asked to engage in a simulated booking of a four-day, three-night trip to New York City. The goal was to negotiate the best price possible for their hotel stay with a digital avatar: either a White AI bot, an Asian AI bot or a Black AI bot. Despite existing research that finds Black people are more likely to be stereotyped as having less competence compared to White or Asian people, the researchers find that the Black AI bot was characterized with higher competence than Asian or White bots. Participants also described the Black bot as more human. Combined, these traits led to higher ratings of satisfaction.
As more businesses consider using AI agents as the front door to their goods and services, the researchers say these findings may play a powerful role in shaping human perceptions and behaviors.
WATCH: Learn more about the key findings in this animated video of the study.
The paper, “I’m Only Human? The Role of Racial Stereotypes, Humanness, and Satisfaction in Transactions with Anthropomorphic Sales Bots,” was published in the Journal of the Association for Consumer Research. You can find the paper here. Researchers from the University of Georgia, Bentley University and the University of Illinois at Urbana-Champaign co-led the study.
If you would like to speak with Professors Perry or Olsen, please contact GW Media Relations Specialist Cate Douglass at cdouglass@gwu.edu.
