CADi · Financial-crime alert decisions
Can you explain why that alert was closed?
Give investigators a clear record of why an alert was escalated, reported or closed, with the reasoning and alternatives available for review.
Bring us a problem, a suspected opportunity or a result that does not add up. We can explore how CADi AML could help, starting with your question and the evidence available.
Get in touchFinancial-crime teams close more alerts than there are hours to close them well. Under that pressure, a suspicious-activity alert gets dispositioned - SAR or no-SAR - to whoever had time, at whatever standard the day allowed.
When a closed alert is later questioned, the firm needs to show how the disposition was reached, under what policy, on what reasoning. A backlog closed inconsistently is a backlog that cannot be defended alert by alert.
CADi AML evidences how a suspicious-activity alert was dispositioned - SAR or no-SAR - alongside the firm's own monitoring. The MLRO still makes the call; CADi attests how it was reached.
It produces a per-alert record of the policy, the reasoning and an immutable trace, so an investigations backlog closes to a consistent, defensible standard rather than to whoever had time. It enters as evidence and is designed to graduate toward governing the alert decision.
A per-alert record of the policy and the reasoning behind the SAR/no-SAR call - consistent quality the MLRO can stand behind.
The backlog closes to one defensible standard, rather than to whoever had time and whatever the day allowed.
When a disposition is challenged, evidence of how it was reached - without reconstructing the investigation from logs and memory.
Runs alongside the firm's monitoring first, attesting how alerts were dispositioned; graduates toward governing the alert decision as trust builds.
Investigations functions closing high alert volumes, who must dispose of each one to a standard that holds up when it is questioned.
The accountable owner of the SAR/no-SAR call and the assurance function behind it, who need the reasoning evidenced, not just the outcome logged.
CADi AML is one product in the CADi range: the same deterministic causal engine - nine layers, a do-calculus core and a hash-chained audit trail - pointed at the alert-disposition decision.
It belongs to the reporting layer: it runs alongside the firm's own monitoring, evidencing how an alert was dispositioned rather than making the call, and is built to graduate into the decision itself once the evidence loop is trusted - enter on evidence, earn the decision.
CADi is in production - its engines are live in the cloud - with nine UK patent applications pending. CADi AML evidences how an alert was dispositioned; the MLRO still makes the SAR/no-SAR call, and CADi never guarantees a regulatory outcome.
Causal analysis runs on a declared model of the decision environment. Its conclusions depend on that model, the available evidence and explicit assumptions; estimates are analytical approximations. The engine has been tested against injected, synthetic and real ingested data. Live-outcome validation is completed against each client's own outcomes during deployment.