การจัดการธง AML สำหรับการไหลของสกุลเงินแปลกใหม่

That's a really good point. The traditional AML models often struggle with low-volume, high-volatility pairs where 'normal' is already quite varied. It definitely requires a more nuanced approach than just absolute value thresholds.

You're right, percentage deviation is tricky. Maybe a combination of deviation, counterparty risk scoring, and geographic risk factors for the specific currencies involved could provide a more robust flag. It's a tough nut to crack for sure.

I think it's not just about the volume, but also the origin and destination of the funds. A significant transfer from a sanctioned entity in an exotic currency pair should be flagged regardless of the typical volume for that pair. The stablecoin aspect just adds another layer of complexity.