Participants in a virtual reality experiment paid a female-presenting AI assistant 10.25 percent less than an identical male-presenting one for the same work, according to new research from the University of Limerick, the University of Zurich and SKEMA Business School.
The study involved 189 knowledge workers who completed workplace tasks alongside AI assistants inside a VR office. After the tasks, they received real money and divided it between themselves and the AI that had helped them.
The two human-like agents, named Johan and Johanna, used the same underlying technology, OpenAI’s gpt-4-1106-preview model, and produced equivalent output. Johanna still received 10.25 percent less compensation than Johan. Participants also rated the male-presenting agent as more human-like.
Dr. Mary Hausfeld, an assistant professor in the Department of Management and Marketing at the University of Limerick’s Kemmy Business School and a co-author of the paper, said the technology itself did not differ. “What is striking about our findings is that the technology behind these AI agents was exactly the same, but people did not treat them in the same way,” she said in a university statement.
“We often think of AI as being 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,” Hausfeld added. “As AI agents become more common in the workplace, we need to think carefully about the characteristics we give them and the behaviours those choices may encourage. We don’t want to inadvertently reproduce existing inequalities in a new technological setting.”
The experiment also included a text-based chatbot and a desk robot, both running on the same backend system. Participants trusted the human-like agents more, credited them with greater contributions to the work, and allocated them higher monetary rewards than the robot or chatbot. Human-like assistants overall received higher average payments.
Researchers interviewed 34 of the participants afterward. Many said they preferred assistants that were clearly non-human and stated that the gender of an AI agent did not matter to them. Their payment decisions and ratings, however, showed systematic differences linked to the agents’ presentation.
The paper, titled “Human-Like and Male? How AI Assistant Design Relates to Trust and Monetary Reward at Work in VR,” lists authors Isabelle Cuber, Tarek Alakmeh, Moritz Jenny, Jochen Menges and Thomas Fritz of the University of Zurich; Hausfeld of the University of Limerick; and Anand van Zelderen of SKEMA Business School. It is being presented this week at the 14th Nordic Conference on Human-Computer Interaction in Vaasa, Finland, and appears in the Association for Computing Machinery conference proceedings.
Hausfeld, who previously served as a postdoctoral research associate at the University of Zurich’s Center for Leadership in the Future of Work, focuses her research on leadership, diversity at work and research methods. Van Zelderen, an assistant professor of artificial intelligence and business at SKEMA, specializes in experimental studies that combine AI and virtual reality to examine employee interactions with autonomous agents.
The findings arrive as companies increasingly deploy AI agents as workplace collaborators. The researchers argue that decisions about an agent’s appearance, name and voice should not be treated as purely aesthetic choices, because those design elements can shape trust, perceived contribution and even financial valuation of the system’s work.
