Google Cloud Technical Consultant: LLM Evaluators Carry Systematic Biases, Proposes Three-Pillar Statistical Audit
Héctor González Álvarez, Senior Technical Solutions Consultant at Google Cloud, argues that LLM evaluators carry their own biases: position bias can flip verdicts over 75% of the time, and verbose, confident-sounding answers receive high scores even when factually wrong. He proposes a three-pillar audit method—multi-pass consistency checks for internal stability, chance-corrected metrics for human alignment validation, and decomposed prompting with multi-model ensembles to avoid the halo effect—offering teams using LLM-as-a-judge an actionable path to assess evaluator quality.
ImportanceMaterialEvidenceE3 inspectableWrite-upStandard
There is a common misconception in tech: human reviewers are subjective, so we should rely on AI for objective evaluation. Héctor González Álvarez, Senior Technical Solutions Consultant at Google Cloud, directly challenges this premise in a consulting blog post published in October 2026. His core argument: LLM evaluators (LLM-as-a-judge) are not objective arbiters but models carrying their own systematic biases; if their quality is not audited, work outputs may be compromised.
The article enumerates five biases documented in research, two of which come with specific magnitudes. Position bias: swapping the order of two options causes an LLM judge to flip its verdict over 75% of the time. This means that if an evaluation pipeline fixes the arrangement order of candidate answers, the scoring results are inherently unreliable. Verbosity bias: evaluators tend to give higher scores to longer, more confident-sounding answers even when the underlying facts are wrong. Additionally, there is the halo effect (clean formatting masks content errors), agreeableness bias (failure to catch incorrect or unsafe outputs can reach 75%), and self-enhancement bias (evaluators systematically overrate outputs generated by themselves or from their own model family).