- Who it helps
- Researchers and professional teams designing task-specific AI evaluations.
- Availability
- The August 23 paper is public and reports acceptance at AIES 2026. A ready-to-use evaluation product was not verified.
Build criteria with practitioners
Nineteen social workers took part in eight workshops. Their participation shaped the evaluation criteria and challenged assumptions about useful model behavior.
Keep the evidence in scope
The authors report agreement between worker assessments and an AI judge, and differences across six models. This does not establish improved outcomes for the people receiving services.
What you can do
- Read how participants created and applied the criteria before adapting the method to another profession. Open reference ↗
Where the evidence stops
The study covers one organizational setting. No workshop or evaluation was reproduced here.
Sources read. Sources checked October 21, 2026. No product workflow or benchmark was executed.
Sources & further reading
Paper · paper
Worker-driven evaluation of AI with school social workers ↗
Selected for direct evidence about the subject, its availability, limitations or the provenance of discussion.
Worker-driven evaluation paper authors · page checked 2026-10-21 · Published 2026-08-23
Coverage note: Independent hands-on testing and cross-platform discussion are incomplete. Public X search failed; no missing links or engagement counts are inferred.
The record over time (1)
Dated source checks and published explanations. An observation date can differ from the original release date.
- · Added to Current
An August study involves school social workers in defining how AI-generated professional work should be evaluated.