EU AI Act requirements (Articles 9–15) are policy and process attestations that Admina cannot infer from runtime traffic alone — they require a human to declare which checks the organisation has actually completed (e.g. "risk management process documented", "data bias examination performed").

Until you submit a real assessment, the dashboard renders all articles at 0% with a default placeholder. To record your real coverage, POST your declared evidence to /api/compliance/gap-analysis.

Example request:

curl -X POST http://localhost:8080/api/compliance/gap-analysis \
  -H "X-API-Key: $ADMINA_API_KEY" -H "Content-Type: application/json" \
  -d '{"risk_category": "high",
       "current_compliance": {
         "risk_management":         [true, true, false, true],
         "data_governance":         [true, false, true, true],
         "technical_documentation": [false, false, true, false],
         "record_keeping":          [true, true, true, true],
         "transparency":            [true, false, false, false],
         "human_oversight":         [false, true, false, false],
         "accuracy_robustness":     [true, true, false, true]
       }}'

Each list has 4 booleans, one per check defined for that article. Replace the placeholder values with your real evidence. After submitting, refresh the dashboard to see the article grid and Admina Score reflect your actual coverage.