CADi · Financial crime
Suspicious-activity and SAR/no-SAR evidence, so an investigations backlog closes to a defensible standard.
How the alert was dispositioned - closure quality an MLRO can stand behind.
Seeking build partnersCADi AML is an open invitation: we are looking for partners to build it with us. If you run financial-crime investigations and want closure quality evidenced, we would like to hear from you.
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 at v0.9 - built and running end-to-end - on nine pending UK patents. CADi AML evidences how an alert was dispositioned; the MLRO still makes the SAR/no-SAR call, and CADi never guarantees a regulatory outcome.
The causal mathematics is rigorous and correctly implemented. Today it runs on an assumed causal graph with heuristic inputs; it is analytical approximation on assumed inputs, not yet statistically validated against live outcomes. We lead with the rigour, and we name the caveat.