Key features include:
Added detailed error insights for failed AI evaluations, enabling supervisors and administrators to understand failure causes and take appropriate corrective actions.
Improved bulk evaluation monitoring by introducing a Records Failed indicator in run history and displaying error details for failed records within evaluation runs.
Enabled supervisors to retry failed evaluations directly from both the evaluation grid and bulk evaluation run lists.
Business Value:
This feature improves troubleshooting by providing supervisors with clear, actionable error messages that explain what went wrong and the recommended next steps. It also includes retry functionality, allowing supervisors to rerun failed evaluations.
Enabled for:
Enabled for end users automatically
GA date: July CY2026