"So There's a Catch-22 Here": How Early Adopters Who Build Multi-Agent LLM Systems Conceptualize Transparency

2026-06-06

AI Summary

This study addresses the conceptual ambiguity surrounding transparency in multi-agent large language model (LLM) systems, where existing definitions fail to account for their distributed architectures and coordination complexities. Through semi-structured interviews with 13 early-stage developers, combined with thematic analysis and insights from human-computer interaction and responsible AI research, the work proposes the first multidimensional transparency framework tailored to multi-agent LLM systems. The framework reconceptualizes transparency as an embodied sociotechnical practice oriented toward developers, end users, and governance stakeholders. It identifies five complementary perspectives—reproducibility, debugging, boundary setting, visualization, and auditing—that collectively offer both theoretical grounding and practical guidance for the design of future AI systems.

Abstract

Multi-agent large language model (LLM) systems are rapidly emerging, yet transparency, a cornerstone of responsible AI, remains under-defined in these distributed architectures, which have complexities of inter-agent coordination and orchestration. In this paper, we present one of the first empirical study of how early adopters of multi-agent LLM systems, who are both the builders and users, understand and practice transparency. We conducted semi-structured interviews with 13 early adopters in [Large Technology Organization] and applied thematic analysis to identify recurring patterns. Participants articulated divergent yet complementary framings of transparency, including reproducibility, debugging, boundary-setting, visualization, and auditing. These perspectives spanned questions of what transparency entails, why it matters, and how it is achieved. We synthesize these into a multidimensional framework, which is developer, user, and governance-focused positioning transparency as a situated socio-technical practice that informs future HCI and AI design and research around aligning expectations and capacities of their intended audiences.

Problem

• transparency
• multi-agent LLM systems
• responsible AI
• distributed architectures
• socio-technical practice

Innovation

• multi-agent LLM systems
• transparency
• empirical study
• socio-technical practice
• thematic analysis