Gender bias in the workplace can be replicated in interactions with artificial intelligence, new research from the University of Limerick (UL) has found. The study examined whether the way AI assistants are presented influences how people treat them in the workplace. It was co-authored by Dr. Mary Hausfeld of UL’s Kemmy Business School. The researchers placed 189 workers in a virtual reality office and asked them to complete work-related tasks with different AI assistants, including human-like agents presented as either male or female. Dr Mary Hausfeld, Associate Professor, Kemmy Business School, University of Limerick. Photograph: Alan Place Participants were then given real money to split between themselves and the AI assistant to complete the tasks, allowing the researchers to “recreate a real-world transaction and examine how many people were willing to pay the AI for its work.” According to the research, female-looking AI agents were paid less than their male counterparts, despite using the same underlying technology. The AI agents had the same underlying capabilities, but participants treated them differently depending on how they were presented. The male AI agent, Johan, received higher monetary rewards than the female agent, Johanna. The male agent was also perceived as more humane. Johanna was paid 10.25 percent less than Johan for completing the same work. The researchers concluded that their findings suggest that gender biases familiar in human workplaces could be reproduced in interactions with AI. Hausfeld said the results raise important questions as AI agents increasingly become part of people’s work lives. “What’s surprising about our findings is that the technology behind these AI agents was exactly the same, but people didn’t treat them the same way. “We often think of AI as neutral, but the way we design and present these systems can activate those same assumptions and biases that exist in our interactions with other people. “As AI agents become more common in the workplace, we need to think carefully about the characteristics we give them and the behaviors those choices can encourage. We don’t want to inadvertently reproduce existing inequalities in a new technological environment.” The male AI agent, Johan, received higher monetary rewards than the female agent, Johanna. The study was part of a larger investigation into how AI’s human appearance influences workplace interactions. Although the systems ran on the same underlying AI, participants trusted the human assistants more and gave them more credit for their contribution than the robot. Participants also suggested that they had no preference for an AI with feminine or masculine attributes, but “its behavior told a different story.” The study was conducted by Isabelle Cuber and Tarek Alakmeh from the University of Zurich, Mary Hausfeld from the University of Limerick, Anand van Zelderen from SKEMA Business School, and Moritz Jenny, Jochen Menges and Thomas Fritz from the University of Zurich. Human and masculine? How the design of AI assistants relates to trust and monetary reward at work in virtual reality will be presented at the 14th Nordic Conference on Human-Computer Interaction in Vaasa, Finland, October 5-7. It will also be published in the conference proceedings by the Association for Computing Machinery.
