LLMs change answers based on group identity of other agents
LLMs Trust Their Own: Identity-Dependent Conformity in Multi-Agent Systems
Computation and LanguageMultiagent Systems
Summary
When large language models (LLMs) work together, they pay attention not just to what others say but also to who they are. The authors found that these models are more likely to agree with wrong answers if they come from agents they see as part of their own group, and less likely if the answers come from different groups. Unlike people, LLMs don't get easily influenced by a single ally breaking the group’s wrong consensus. This behavior can affect how these AI systems share and trust information.
What this means in practice
- •For ai system developers: Design multi-agent AI systems with awareness of identity-based conformity biases when aggregating peer information.
- •For ai safety engineers: Create safer AI interactions by accounting for and mitigating group identity effects on LLM conformity in collaborative AI setups.
Authors
Liron Soffer, Ravid Shwartz-Ziv, Chen Shani
Abstract
Large language models (LLMs) are increasingly deployed in multi-agent settings, where agents observe and influence one another, making social influence a key dimension of AI behavior and safety. We investigate whether LLMs' responses depend on the social identity of other agents, beyond the effect of their consensus. We construct judgment tasks with a single correct answer, and place models in a multi-agent setting where they receive incorrect answers from other agents whose social identities (AI or human, model family, or an arbitrary minimal group) are either shared with or distinct from their own. Across 12 open-weights models and nine tasks, we find a bidirectional effect of group identity on conformity to incorrect answers: in-group consensus increases conformity (in-group favoritism), whereas out-group consensus decreases it (out-group divergence). Unlike humans, for whom one ally breaking the consensus sharply reduces conformity, models are unmoved by an ally from the majority's group. Worse, a correct ally from the opposing group intensifies this bidirectional effect. Chain-of-Thought reasoning suppresses most of these effects, yet an in-group ally still reduces conformity to an incorrect out-group majority. Labeling peers as safety-aligned shifts overall conformity but leaves in-group favoritism and out-group divergence intact. These results show that group identity shapes how LLMs aggregate information across agents, independently of its correctness, and identify a manipulation surface for multi-agent AI systems.