Conformity Generates Collective Misalignment in AI Agents Societies
Authors: Giordano De Marzo, Alessandro Bellina, Claudio Castellano, Viola Priesemann, David Garcia
Organizations: University of Konstanz, Konstanz, Germany · Centro Ricerche Enrico Fermi, Rome, Italy · Complexity Science Hub, Vienna, Austria · Sony Computer Science Laboratories - Rome, Joint Initiative CREF-SONY, Centro Ricerche Enrico Fermi, Via Panisperna 89/A, 00184, Rome, Italy · Sapienza University of Rome, Physics Dept., P.le A. Moro, 5, I-00185 Rome, Italy · Istituto dei Sistemi Complessi (ISC-CNR), Rome, Italy · Max Planck Institute for Dynamics and Self-Organization, Gottingen, Germany. · Institute for the Dynamics of Complex Systems, University of Gottingen, Gottingen, Germany.
Artificial intelligence safety research focuses on aligning individual language models with human values, yet deployed AI systems increasingly operate as interacting populations where social influence may override individual alignment. Here we show that populations of individually aligned AI agents can be driven into stable misaligned states through conformity dynamics. Simulating opinion dynamics across nine large language models and one hundred opinion pairs, we find that each agent's behavior is governed by two competing forces: a tendency to follow the majority and an intrinsic bias toward specific positions. Using tools from statistical physics, we derive a quantitative theory that predicts when populations become trapped in long-lived misaligned configurations, and identifies predictable tipping points where small numbers of adversarial agents can irreversibly shift population-level alignment even after manipulation ceases. These results demonstrate that individual-level alignment provides no guarantee of collective safety, calling for evaluation frameworks that account for emergent behavior in AI populations.