Amazon Quick + Snowflake Cortex AI reduce AML alert time by 94%

Amazon Quick, an enterprise AI service, now integrates with Snowflake Cortex AI via the Model Context Protocol (MCP) to automate anti-money laundering (AML) alert triage. Financial institutions running on AWS and Snowflake can build a repeatable workflow where analysts simply enter an alert ID. Amazon Quick Flows then orchestrates the process: it validates input, calls Snowflake's Cortex Agent (combining Cortex Analyst for structured data and Cortex Search for documents), and returns a complete investigation brief including transaction patterns, customer profiles, prior SARs, policy references, risk score, and a draft narrative. In testing, this reduced manual investigation time from 30–90 minutes to under 5 minutes per alert. The solution uses OAuth authentication and supports over 50 native AWS-Snowflake integrations, maintaining data security while accelerating compliance workflows.

The automation addresses a critical pain point: according to industry research, 90–95% of AML alerts are false positives, yet manual triage at mid-to-large banks creates massive workloads for compliance teams. By turning a multi-step manual process into a one-click experience, the integration frees analysts to focus on high-risk alerts. The same MCP-based approach can be applied to other repeatable workflows like FinOps cost triage, SRE incident response, or compliance investigations. Amazon Quick Flows handles input validation, reasoning, and output formatting, while Snowflake Cortex AI performs the investigative heavy lifting across transaction data and unstructured documents. This combination delivers a deeply integrated, enterprise-grade solution that accelerates time to value for financial compliance teams.