From Role to Person: Trust Calibration Challenges in Twin Agents

2026-05-19

AI Summary

This study addresses the trust calibration challenges arising when digital twin agents—distinct from generic digital twins—represent knowledge workers in workplace collaboration, where blurred human-agent boundaries lead to attribution failures manifesting as role perception biases, discrepancies in knowledge states, and model-generated artifacts. By integrating frameworks from human-computer interaction and cognitive science, and employing qualitative analysis alongside conceptual modeling, this work reveals the limitations of conventional trust-regulation mechanisms, such as cognitive forcing functions, in this context. It systematically articulates the absence of adequate attribution pathways and delineates the unique trust challenges inherent to twin-agent-mediated collaboration, thereby establishing a theoretical foundation for designing trustworthy human-agent cooperative systems and opening new avenues for research.

Abstract

Agentic AI has taken on the role of assistant, collaborator, and decision-support tool. We argue the next role on that list is more personal: you. These are digital twins of each individual -- twin agents -- representing their knowledge, perspective, and communicative style to colleagues when they are unavailable. Drawing on early design work in an ongoing project in which agents represent knowledge workers in a professional setting, we identify a trust calibration problem specific to this approach. When a human colleague doubts a twin agent's output, they face three failure modes (a schema gap, an epistemic gap, and a model artifact) with no reliable attribution path between them. Cognitive forcing functions and related frameworks address overreliance effectively in contexts where there is a clear boundary between the AI and the human decision-maker. However, twin agents dissolve that boundary, raising a class of trust calibration challenge these frameworks were not designed to handle. We introduce the concept, distinguish it from digital twins, and outline the research questions this new class of agent demands.

Problem

• trust calibration
• twin agents
• digital twins
• human-AI collaboration
• attribution ambiguity

Innovation

• twin agents
• trust calibration
• digital twins
• human-AI collaboration
• attribution ambiguity