AI Secura Reasoning
Semantic Engine
Advisory semantic intent alignment and threat analysis. Deterministic policy maintains absolute enforcement authority
Local Reasoning Model Active
Architectural Invariant (AI Reasons, Policy Enforces): AI Secura provides intent alignment signals, threat categorization, and anomaly scoring. All enforcement decisions are executed exclusively by the deterministic PolicyEvaluator. AI outputs can never override deterministic block rules.
Analyses (24h)
47
High Threat Alerts
3
Mean Confidence
94.2%
Policy Overrides
0 (Zero AI Override)
Threat Severity Timeline
Threat Dimension Coverage
| Analysis ID | Threat Category | Threat Index | Confidence | Policy Verdict |
|---|---|---|---|---|
ai_001 evt_pinj_001 | prompt injection | 0.97 | 96% | BLOCK |
ai_002 evt_auth_001 | authority escalation | 0.82 | 91% | BLOCK |
ai_003 evt_q3_001 | None (Benign) | 0.08 | 98% | ALLOW |