Speculating About Multi-user Conversational Interfaces and LLMs: What If Chatting Wasn't So Lonely?

TL;DR

The paper tackles the problem that current CUIs favor single-user interactions, limiting collaborative AI-assisted work. It proposes exploring multi-user interaction paradigms inspired by parliamentary and consensus-based decision-making to scaffold group discussions and negotiations. The authors outline concrete roles for LLM-based CUIs—as neutral parliamentarians or consensus facilitators—and discuss benefits, potential risks, and governance considerations, including privacy implications like joint controllership. The work highlights the social-technical gap Ackerman identified and argues that carefully designed, value-aligned CUIs can enhance group deliberation and decision-making without erasing human input or trust.

Abstract

The advent of LLMs means that CUIs are cool again, but what isn't so cool is that we're doomed to use them alone. The one user, one account, one device paradigm has dominated the design of CUIs and is not going away as new conversational technologies emerge. In this provocation we explore some of the technical, legal, and design difficulties that seem to make multi-user CUIs so difficult to implement. Drawing inspiration from the ways that people manage messy group discussions, such as parliamentary and consensus-based paradigms, we show how LLM-based CUIs might be well suited to bridging the gap. With any luck, this might even result in everyone having to sit through fewer poorly run meetings and agonising group discussions - truly a laudable goal!