New publication in MIS Quarterly
André Hanelt, Manuel Wiesche, Alexander Benlian, Oliver Hinz (2025) "Opening the Network of Trust: How Domain Experts in Triadic Relationships Build Trust in AI-Based Counterparts", forthcoming https://doi.org/10.25300/MISQ/2025/18041
The study examines how trust is built between domain experts and AI-based counterparts in triadic relationships with customers. Using data from the introduction of the robo-advisor EVA at EuroBank, we examine how financial advisors deal with trust issues between opacity and performance as well as disconfirmation and complementation. Through relational interpretation (categorizing the opportunities and risks of AI in their relationship networks) and relational adaptation (adapting these networks in response), subject matter experts go through three different phases of building trust. First, they avoid vulnerabilities by emphasizing the imperfections of AI and preventing customer interactions. Then they secure vulnerabilities through human support mechanisms and controlled integration of the AI. Finally, they accept vulnerabilities by experimenting with the robo-advisor themselves and hybridizing customer relationships. This tripartite perspective shows how social contexts influence and shape trust formation and goes beyond technology-centric approaches to consider relational dimensions of trusting beliefs and behaviors in AI adoption.
The journal "Management Information Systems Quarterly" (MIS Quarterly) is considered one of the most influential journals in the field of information systems and sets the standard for rigorous research methodology and theoretical contributions.