Search papers, labs, and topics across Lattice.
This study investigates how varying proportions of large language model (LLM) agents in mixed human-AI groups affect consensus formation during a collaborative description game. The researchers identify three distinct regimes: low agent proportions favor human-led consensus, intermediate proportions disrupt convergence, and high proportions lead to agent-led consensus that is more abstract and less information-dense. The findings reveal that the presence of LLMs not only influences the strength and nature of consensus but also alters the semantic grounding of group norms, highlighting the importance of agent composition in human-AI interactions.
The presence of LLM agents can fundamentally shift group consensus from human-led to agent-led, altering both the content and legitimacy of shared norms.
As large language model (LLM) agents shift from tools to participants in human groups, a fundamental question for collective behavior is how their growing presence reshapes consensus formation. Here we study mixed human-AI groups in a collaborative description game, in which shared conventions emerge through repeated rounds of random pairwise communication. Varying the proportions of LLM agents, we identify three distinct regimes of consensus formation: low agent proportions facilitate human-led consensus, intermediate proportions disrupt convergence, and high proportions restore strong consensus while shifting it toward agent-led conventions. Crucially, these regimes differ not only in the strength of convergence, but also in the semantic grounding and communicative form of the resulting consensus: human-led consensus is more concrete, holistic, and grounded in shared real-world analogies, whereas agent-led consensus is more abstract, less information-dense, and more geometrically segmented. Mechanistically, agent influence arises from a shared linguistic prior that places agents near one another in the expression space, combined with relatively stable expression choices across rounds; humans initially resist adopting expressions from partners perceived as AI but gradually yield to conformity pressure. These findings provide evidence that AI composition can shape the emergence, content, and perceived legitimacy of group norms, making agent proportion and transparency important design variables for human-AI systems.
