How ActLoom Works
The platform follows the actual lifecycle of AI compliance. You first define what systems exist, then determine which obligations apply, then measure your current level of readiness, then produce evidence and maintain it over time. That sequence matters because later modules depend on the accuracy of the earlier ones.
Platform logic from first login to operational compliance
Map
Register each AI system with enough context to understand purpose, users, autonomy, and deployment scope.
Assess
Classify risk, evaluate controls, and identify the obligations that are met, partial, or missing.
Prove
Turn the current state into reports, technical records, and shareable evidence for customers, audits, or regulators.
Operate
Coordinate owners, governance reviews, incidents, and post-market monitoring so compliance stays alive after launch.
| Platform stage | Main question answered | Primary output |
|---|---|---|
| AI Systems | What AI do we have, and why does it exist? | Structured inventory with system scope, type, and classification context |
| Compliance | What obligations apply, and where are the gaps? | Score, gap analysis, remediation plan, and evidence-backed answers |
| Documents | How do we package our current posture into something shareable? | Reports, technical documentation, transparency notices, and versioned exports |
| Governance / Monitoring / Incidents | How do we keep the programme credible over time? | Audit trail, review cadence, event logs, incident records, and continuous follow-up |