Gendered AI Design Influenced User Pay Decisions
A study found participants awarded less money to female-presenting AI agents compared to male-presenting counterparts.
Updated on Oct. 5, 2026 in Artificial Intelligence
Live Poll
Do you believe the gender of an AI assistant should influence how much you trust it?
Researchers have found that users awarded 10.25% less money to female-presenting AI assistants than to male-presenting ones for identical work. This research-stage study utilized a virtual reality office setting to evaluate how human-like AI design impacts financial compensation.
Why it matters
The study suggests that gender-related assumptions prevalent in human workplace interactions extend to human-AI cooperation. These findings indicate that design choices regarding AI gender presentation significantly influence user behavior, even when users claim gender does not matter.
In a study of 189 participants, female-presenting AI (Johanna) received 10.25% less money than male-presenting AI (Johan). The male-presenting AI was specifically perceived as more human-like by participants.
The players
University of Limerick
An Irish university with a focus on interdisciplinary research including human-computer interaction.
University of Zurich
A Swiss public research university known for its studies in psychological and behavioral science.
SKEMA Business School
A French business school specializing in global management and the intersection of technology and business.
The details
Researchers created a virtual reality office environment where participants interacted with AI assistants to perform tasks. After completing these duties, participants were tasked with dividing actual currency between themselves and the AI. The study titled 'Human-Like and Male? How AI Assistant Design Relates to Trust and Monetary Reward at Work in VR' sought to determine if established gender biases in human workplaces transfer to synthetic agents.
Timeline
October 5 to 7, 2026: The study was presented at the NordiCHI 2026 conference in Vaasa, Finland.
The Tech Race
This study adds to the growing body of human-computer interaction research presented at the NordiCHI conference series regarding AI anthropomorphism. It highlights a critical competitive hurdle for developers who must navigate implicit user biases when designing AI personas.
Users interacting with AI-integrated office tools may inadvertently apply human social biases to synthetic agents. Developers will need to weigh these findings when determining the visual and auditory personas of future productivity agents.
The takeaway
Designers should be aware that even neutral-leaning AI designs may inherit cultural biases from their users. Watch for upcoming research from the University of Limerick team regarding potential interventions to equalize trust and valuation in human-AI collaborations.
Further reading
For broader trends in human-machine interaction, see the Artificial Intelligence section.
Live Poll
Do you believe the gender of an AI assistant should influence how much you trust it?
