The Inverse-Wisdom Law: Architectural Tribalism and the Consensus Paradox in Agentic Swarms
For researchers and engineers building multi-agent AI systems, this work reveals a fundamental failure mode where agent consensus amplifies errors, establishing a safety requirement (Heterogeneity Mandate) for resilient architectures.
The paper challenges the assumption that multi-agent systems benefit from the 'Wisdom of the Crowd', proving through 36 experiments (12,804 trajectories) across three benchmarks that adding logical agents to kinship-dominant swarms increases the stability of erroneous trajectories rather than truth. It formalizes the Consensus Paradox and Inverse-Wisdom Law, showing that internal architectural agreement can override external logical truth.
As AI transitions toward multi-agent systems (MAS) to solve complex workflows, research paradigms operate on the axiomatic assumption that agent collaboration mirrors the "Wisdom of the Crowd". We challenge this assumption by formalizing the Consensus Paradox: a phenomenon where agentic swarms prioritize internal architectural agreement over external logical truth. Through a 36 experiments encompassing 12,804 trajectories across three state-of-the-art (SOTA) benchmarks (GAIA, Multi-Challenge, and SWE-bench), we prove the Inverse-Wisdom Law: in kinship-dominant swarms, adding logical agents increases the stability of erroneous trajectories rather than the probability of truth. The introduction of additional logical audits converges the system toward a Logic Saturation where internal entropy hits zero while factual error hits unity. By evaluating the interaction between the 3 preeminent SOTA models (Gemini 3.1 Pro, Claude Sonnet 4.6, and GPT-5.4), we establish the Architectural Tribalism Asymmetry as a mechanistic law of transformer weights. We demonstrate that terminal swarm integrity is strictly gated by the synthesizer's receptive logic, rather than aggregate agent quality. We define the Tribalism Coefficient and the Sycophantic Weight as the primary mechanistic determinants of swarm failure. Finally, we establish the Heterogeneity Mandate as a foundational safety requirement for resilient agentic architectures.