Human-led compliance governance.
AI drafts, people decide.
One ledger, from regulator to ratified decision.
- SourceRegulatory sourceJFSC · FCA · GFSC
Live regulator feeds ingested and classified before they reach the obligation graph.
- ObligationObligation272 in scope
Each rule mapped to a named obligation with owner, scope and coverage state.
- RiskRisk & appetite260 at risk · 12 green
Residual rating compared to appetite — breaches escalate, in-band moves are bounded.
- ControlPolicy & control0 in review
Versioned policies and the controls that operationalise them — every change tracked.
- TestEvidence & testRisk-stratified sampling
Tests are planned with documented methodology; evidence is hashed and retained.
- FindingFinding & issue8 open · 8 issues
AI drafts severity and rationale; a named human confirms, overrides or closes.
- DecisionDecision & ratification6 pending ratification
Every decision sealed into the tamper-evident record before the board sees it.
Cohort 1 Source-of-Wealth — JFSC Handbook §4.6 → residual risk elevated +6pts → control CTL-SOW-022 uplift → re-test TEST-2025-331 → 8% defect rate → finding F-2025-241 → policy amendment POL-AML-04 §6 tabled for Board ratification 18 Dec.
42 relationships in scope · pending Risk Committee 04 Dec · every step in the chain.
RegAlign preserves compliance skill on purpose. The AI drafts; the named human still reasons. Every obligation, control and decision surfaces a Why this recommendation panel — what the regulator says, what the AI suggests with its confidence, the questions you (the CCO) need to answer, and one line where the AI argues against itself. You get the questions, not just the answer — so the function keeps its technical ability rather than atrophying into click-approval.
RegAlign® supports compliance governance. It does not provide legal or regulatory advice. Decisions, approvals and overrides remain the responsibility of identified humans in the audit trail.