LiveEx Shield's Fraud Framework maps line-for-line to the recommendations published by FATF, Wolfsberg, Basel, the EU and your home regulator. Every alert, case and report it produces is shaped the way examiners and auditors already expect — no custom mapping layer, no reconciliation work.
The LiveEx Shield Fraud Framework runs on the same real-time scoring backbone as the AML engine — and writes into the same alert queue, evidence store and audit trail. Three principles describe how it works in production.
Every authorisation, login and payment passes through configurable fraud rules, behavioural baselines and anomaly scoring before the customer sees a response. Scenario packs ship for account-takeover, APP scams, card-not-present misuse, synthetic identity, mule networks and internal fraud — tunable in production without a code release.
Static blocklists are obsolete the day they ship. The framework builds a per-customer behavioural baseline — typical counterparties, volumes, devices, geographies — and flags statistically significant deviations. Investigators see the baseline, the deviation and peer-group comparisons side-by-side. Decisions are explainable to customers, regulators and chargeback boards.
Fraud and AML alerts land in the same queue, against the same customer, with the same evidence store. An analyst working a fraud case sees related AML alerts; an MLRO reviewing a suspicious transaction sees the fraud signals that fired alongside it — the model regulators ask for, and the model that finds the cases neither team would find alone.
The same rule packs, behavioural baselines and audit logic run synchronously inline with the payment flow, asynchronously over batch and historical files, and continuously inside the investigation workbench — with one common alert queue, evidence store and audit trail behind all three.
Every authorisation, transfer, payment and login passes through the framework before the customer sees a response. Sub-second scoring decides allow, step-up, hold or decline from behavioural baseline, scenario rules and live sanctions overlay.
Run the same rule packs across a full transaction history, EOD batch file or any specified date range. Used for periodic reviews, post-incident look-backs, regulator-requested replays and detection-coverage validation on new scenarios.
Alerts from both modes land in a unified queue. Investigators work cases inside the workbench, escalate under 4-eyes review, file SAR / STR / CTR straight from the evidence record, and MI dashboards roll up to management and the board on a fixed cadence.
The Fraud Framework is not a separate product bolted onto the platform. It shares customer profile, screening verdict, transaction record, evidence store and audit trail with the AML modules in real time — so every fraud signal is enriched with AML context and every AML alert benefits from fraud telemetry.
Sanctions, PEP and adverse-media hits attach to fraud alerts automatically.
Transaction Monitoring scenarios share rule packs and behavioural baselines.
Customer risk score, beneficial owners and IDV evidence available inline.
Inherent + residual risk re-rated whenever fraud activity changes the picture.
Same AI oversight layer scoring fraud alerts on match quality and risk severity.
One SAR / STR pipeline. Fraud and AML filings produced from a single evidence record.
