video1.0<iframe src="https://www.loom.com/embed/d263c43c05924d37b3728909b0ac7f10" frameborder="0" width="1110" height="832" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>8321110Loomhttps://www.loom.com8321110https://cdn.loom.com/sessions/thumbnails/d263c43c05924d37b3728909b0ac7f10-ea3ba35b0ee53cfa.gif169.551This Loom explains how Viber’s Ridby uses AI to detect logistics invoice fraud that is hard to catch in standard ERP systems. The CEO notes there are many invoicing variations across hundreds of customers, and fraud can look like a normal invoice with modular elements, creating a roughly half billion dollar a year problem, including a cited example of a fraudulent $3.5 million invoice for a one-inch deed in Connecticut tied to Amazon. He describes a workflow that first checks against the management system, then uses web-sourced verification plus AI search to confirm data authenticity. A demo invoice is sent via Composio into email and Slack, where it is flagged, and prior instances are identified using MediRoute to show it had been seen before.