Trust: AI and data governance
How we keep AI under human control and away from the critical data path.
Autonomy levels
- T0 — observe and report: no action.
- T1 — recommend: a human approves (MLEO quarantine with supervisor approval).
- T2 — pre-approved playbook: actions tested in advance, with a circuit-breaker.
- T3 — autonomous on an allowlist: only blocks of known indicators, with automatic rollback.
The CYBER3.AI copilot works at T0–T1: it explains and recommends; automated actions belong to FasterUp SOC and follow the levels above.
Safeguards
- Human-in-the-loop on any destructive action.
- Kill-switch available to the client.
- Immutable journal: every decision has an ID, source, reason and time.
- Never on the critical data path: AI components are not inline.
Data and compliance
Account and usage data are hosted in the European Union; traffic encrypted in transit and at rest; role-based access with least privilege. Defensive-only posture. Aligned with the EU AI Act requirements for systems under human oversight and with NIS2 reporting requirements. Details: Security & Compliance.
Write to us: [email protected]
Ecosystem
- CYBER3.AISecurity LLM and threat intelligence
- CYBER3Anti-scam apps, VPN included
- FasterUp SOCManaged SOC & VAS with sensors
- FasterUp OTIndustrial network and SCADA monitoring
All are brands of ROL PORTAL SERVICES SRL · ISO/IEC 27001:2022
← Back to CYBER3.AI