Real-Time Group Dynamics with LLM Facilitation: Evidence from a Charity Allocation Task

2026-05-13

AI Summary

This study investigates the impact of large language models (LLMs) acting as coordinators on consensus formation, fairness in resource allocation, and participant perceptions in real-world group decision-making. Through two incentivized experiments involving 879 participants, we compare real-time text-based discussions among three-person groups tasked with allocating charitable funds, under conditions with and without LLM coordination and across different coordination strategies. Results indicate that while LLM coordination does not significantly enhance group consensus or actual participation equity, it shifts funding allocations toward specific organizations by up to 5.5 percentage points and introduces two key governance risks: “algorithmic steering” and an “illusion of inclusivity.” Participants erroneously perceive discussions as more inclusive and express greater trust in the process, revealing the potential for AI-mediated deliberation to subtly manipulate outcomes and distort subjective experiences.

Abstract

As large language models (LLMs) evolve from single-user assistants to active participants in civic and workplace deliberation, evaluating their effects on collective decision making becomes a governance challenge. We present two empirical studies (N=879) of real-time, text-based group deliberation in an incentive-compatible charity allocation task with real financial stakes ($7,200 USD). Groups of three allocate a donation budget under varying LLM facilitation conditions: Study 1 (N=204) compares three frontier models; Study 2 (N=675) compares facilitator strategies against a no-facilitation baseline. Across both studies, LLM facilitation did not significantly improve group consensus in either study, yet participants consistently preferred facilitated discussion. We additionally identify two governance-relevant risks. First, algorithmic steering: facilitators shifted select charity-level allocations by up to 5.5 percentage points -- directly affecting the final charitable payout -- even when aggregate agreement metrics remained unchanged. Second, an illusion of inclusion: participants cited inclusivity as their primary reason for preferring LLM facilitators, yet neither survey nor transcript-based measures of participation equity improved. Notably, participants reported greater trust in the process under the same conditions where facilitators exerted directional influence on outcomes. Together, these findings show that in AI-mediated group deliberation, perceived procedural improvement can coexist with measurable steering and unchanged participation inequality, motivating evaluation practices that treat collective outcomes, interaction dynamics, and participant perceptions as distinct governance targets.

Problem

• group deliberation
• large language models
• collective decision making
• algorithmic steering
• participation equity

Innovation

• LLM facilitation
• group deliberation
• algorithmic steering
• illusion of inclusion
• collective decision making