Right Order, Wrong Scale: Auditing LLM Judges for Occupational AI Measurement
Factual evidence
What the source reports
Research introducing O*NET-BENCH shows LLM ranking agreement fails to translate into reliable acceptance rates for occupational AI tasks.
Inspect the evidence
- Inclusion basis
- Enterprise AI
- Publisher and source type
- arXiv cs.CL — Computation and Language · RESEARCH
- Published by source
- 5 October 2026
- Collected by OneBench
- 6 Oct 2026, 03:01 UK
Stored source excerpt
arXiv:2610.02492v1 Announce Type: cross Abstract: LLM judges are increasingly used to assess whether AI outputs meet workplace requirements, but agreement on response rankings does not…
Short excerpt from the collected text, not the full source. Use the source link to read it in context.
The factual summary is a OneBench synthesis, not a quotation or independent verification. Collection time is not publication time. Open the source for its full context; related reporting can share the same underlying announcement.
OneBench interpretation
Institutional assessment
So what
Using LLM-as-a-judge to evaluate automated workflow accuracy creates uncalibrated operational risk due to underlying acceptance-rate discrepancies.
Do what
Review model validation protocols with the team responsible for model risk management when using LLM judges for workflow testing.