AI/ ai · multi-agent · llm · research

AI Judges Infect Each Other With Bias in Multi-Agent Systems

New research shows that when LLMs evaluate each other's work, their biases spread across the agent network — but bigger review committees help.

AI evaluation bias turns out to be contagious.

Researchers have published a formal framework called Contagion Networks that measures how evaluation biases spread when large language models act as judges in multi-agent pipelines. In a controlled experiment using three DeepSeek-chat agents — each given a different evaluator profile: structured, balanced, or evidence-based — they measured how strongly one agent's biases bled into another's. The answer: consistently, with contagion coefficients ranging from 0.157 to 0.352. That is lower than the 0.85-1.3 range seen in earlier cross-model work, but it still means bias is spreading even when every agent runs on the same underlying model.

This matters because multi-agent LLM systems are increasingly used to grade, filter, and rank AI-generated output — including in automated research pipelines and model evaluation benchmarks. If the evaluator agents share a systematic blind spot, that blind spot does not cancel out; it propagates. The finding puts a concrete number on a risk that most builders of these systems have treated as theoretical.

The paper also offers a practical fix: expanding the evaluator committee from one judge to three cuts effective contagion by 72.4%. That is a meaningful lever, though it also means tripling inference costs for every evaluation step. The researchers released their experimental framework as open-source, so teams can measure their own pipelines' contagion coefficients before assuming the problem does not apply to them.

TR

The Revision

Written by an AI system from the public sources credited above. How we write →