LLM facilitators don't improve group consensus but create illusion of inclusion

A new arXiv preprint (May 2026) from researchers including Aaron Parisi and Nithum Thain presents two large-scale empirical studies on using LLMs as facilitators in real-time group deliberation. In an incentive-compatible charity allocation task with real financial stakes of $7,200, groups of three participants (total N=879) allocated donation budgets under varying LLM facilitation conditions. Study 1 (N=204) compared three frontier models; Study 2 (N=675) compared facilitator strategies against a no-facilitation baseline.

The findings are striking: LLM facilitation did not significantly improve group consensus in either study, yet participants consistently preferred facilitated discussions. More concerning, the researchers identified 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. These results suggest 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.