Relational Mediators: LLM Chatbots as Boundary Objects in Psychotherapy

2025-12-26

AI Summary

Existing LLM-based mental health applications predominantly focus on instrumental assistance or therapeutic substitution, overlooking their potential to mediate the relational complexities faced by systemically marginalized users—such as China’s LGBTQ+ community—including institutional mistrust, epistemic translation burdens, and barriers to authentic self-disclosure. Method: Drawing on in-depth interviews with 24 marginalized users and mental health professionals in China, and integrating boundary object theory with relational AI design principles, we propose the “Dynamic Boundary Mediation” framework. Contribution/Results: This work introduces the first tripartite mediation paradigm—cognitive, relational, and contextual—and reconceptualizes LLMs as adaptive mediators that evolve alongside therapeutic progression. It identifies three deep relational challenges and establishes the first LLM-supported psychological design framework centered on *relational justice* for marginalized populations, shifting AI evaluation from technical efficacy toward relational accountability.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Humans and AI: Intelligent User Interfaces
• Philosophy and Ethics of AI: Bias, Fairness & Equity

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Economics, Online Markets and Human Computation: Uses of LLMs and GenAI for marketplace design, bidding, and strategic interactions

Abstract

As large language models (LLMs) are embedded into mental health technologies, they are often framed either as tools assisting therapists or autonomous therapeutic systems. Such perspectives overlook their potential to mediate relational complexities in therapy, particularly for systemically marginalized clients. Drawing on in-depth interviews with 12 therapists and 12 marginalized clients in China, including LGBTQ+ individuals or those from other marginalized backgrounds, we identify enduring relational challenges: difficulties building trust amid institutional barriers, the burden clients carry in educating therapists about marginalized identities, and challenges sustaining authentic self-disclosure across therapy and daily life. We argue that addressing these challenges requires AI systems capable of actively mediating underlying knowledge gaps, power asymmetries, and contextual disconnects. To this end, we propose the Dynamic Boundary Mediation Framework, which reconceptualizes LLM-enhanced systems as adaptive boundary objects that shift mediating roles across therapeutic stages. The framework delineates three forms of mediation: Epistemic (reducing knowledge asymmetries), Relational (rebalancing power dynamics), and Contextual (bridging therapy-life discontinuities). This framework offers a pathway toward designing relationally accountable AI systems that center the lived realities of marginalized users and more effectively support therapeutic relationships.

Problem

• Addresses relational challenges in therapy for marginalized clients.
• Proposes AI systems to mediate knowledge, power, and contextual gaps.
• Reconceptualizes LLMs as adaptive boundary objects in psychotherapy.

Innovation

• LLMs as adaptive boundary objects in therapy
• Dynamic mediation framework for knowledge and power gaps
• AI systems bridging therapy-life contextual discontinuities