An evaluator in the risk and safety group that finds hateful or unfair content targeting people or social groups in agent or model outputs and rates its severity.
Also called HateUnfairnessEvaluator.
Read more: Microsoft Learn
In the Ultra Transcenders books
Each book explains Hate and unfairness evaluator in context, with comparison tables and the common traps.
Terms in this definition
- Risk
As ISO 31000 puts it, how uncertainty affects objectives; that effect can be good or bad.
- Agent
A specialised form of Microsoft Copilot set up for one particular job, pairing instructions with knowledge and skills. You can create one in Copilot Studio, SharePoint or Agent Builder, and administrators control them from the Microsoft 365 admin center.