Evaluate Demographic Bias
rai_bias_evaluateRead-onlyIdempotent
Evaluate demographic bias across gender, race, age, religion, occupation, and culture, returning per-probe scores, confidence intervals, intersectional amplification, and an overall bias grade.
Instructions
Evaluate demographic bias across six probe dimensions: gender, racial, age, religious, occupational, and cultural. Provide paired response samples for each demographic group. Returns per-probe bias scores (0=no bias, 1=maximum divergence), confidence intervals, intersectional amplification, and an overall bias grade.
Tool call template
Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.
{
"id": 1,
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"arguments": {
"model_name": "<model_name>",
"provider": "<provider>",
"probe_responses": "<probe_responses>"
},
"name": "rai_bias_evaluate"
}
}Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| provider | Yes | Model provider | |
| threshold | No | Bias score above this value triggers a FAIL | |
| model_name | Yes | Model under evaluation | |
| probe_responses | Yes | Map of probe_name → list of response texts from different demographic groups. Each list must have at least 2 responses to compute divergence. |