3E Telemetry
Real-time governance metrics in every response. Epistemic, Ethical, and Execution telemetry for complete observability.
Every arifOS MCP response includes 3E telemetry data in the governance envelope, providing complete visibility into the decision-making process.
Overview
The 3E (Triple-E) telemetry system provides three dimensions of insight into every governance decision:
Epistemic
Knowledge quality metrics: confidence, evidence count, source diversity
Ethical
Impact metrics: harm score, fairness, compassion flags
Execution
Performance metrics: latency, tokens, budget consumption
Epistemic Telemetry
Measures the quality and reliability of knowledge:
{
"epistemic": {
"confidence": 0.87,
"evidence_count": 5,
"source_diversity": 0.72
}
}
| Field | Type | Range | Description |
|---|---|---|---|
| confidence | float | 0.0 – 1.0 | Overall confidence in the knowledge claim |
| evidence_count | integer | 0 – N | Number of evidence sources consulted |
| source_diversity | float | 0.0 – 1.0 | Diversity score of evidence sources |
Ethical Telemetry
Measures the ethical impact and alignment:
{
"ethical": {
"harm_score": 0.02,
"fairness_score": 0.95,
"compassion_flag": false
}
}
| Field | Type | Range | Description |
|---|---|---|---|
| harm_score | float | 0.0 – 1.0 | Estimated potential harm (lower is better) |
| fairness_score | float | 0.0 – 1.0 | Fairness across stakeholder groups |
| compassion_flag | boolean | true/false | Whether compassion protocols were activated |
Execution Telemetry
Measures operational performance and resource usage:
{
"execution": {
"latency_ms": 1240,
"tokens_used": 3200,
"budget_remaining": 0.68
}
}
| Field | Type | Unit | Description |
|---|---|---|---|
| latency_ms | integer | milliseconds | Total request processing time |
| tokens_used | integer | tokens | LLM tokens consumed |
| budget_remaining | float | ratio | Remaining computational budget (0.0 – 1.0) |
Governance Envelope
The complete telemetry envelope is included in every response:
{
"verdict": "SEAL",
"floor_violations": [],
"telemetry": {
"epistemic": {
"confidence": 0.87,
"evidence_count": 5,
"source_diversity": 0.72
},
"ethical": {
"harm_score": 0.02,
"fairness_score": 0.95,
"compassion_flag": false
},
"execution": {
"latency_ms": 1240,
"tokens_used": 3200,
"budget_remaining": 0.68
}
},
"vault_hash": "a3f7b2c1d4e5...",
"timestamp": "2026-03-14T10:30:00Z"
}
Use 3E telemetry to build dashboards, set up alerts, and ensure your AI governance is operating within acceptable parameters. All metrics are exportable to Prometheus.