Custom agent · Built in-house
Check every contact center call for compliance, not a sample
AutoX, a Thai vehicle title loan provider in the SCBX Group, uses AI built on Azure OpenAI to review all of its contact center calls for regulatory compliance and runs a round-the-clock coaching agent for field loan officers.
What they did
Call recordings from the contact center are reviewed by AI against regulatory requirements instead of by staff sampling a handful. Separately, a coaching agent on Azure AI Foundry answers field loan officers' questions at any hour, and Azure AI Vision handles document classification, signature checks and vehicle collateral valuation. Some of these capabilities were co-developed with Microsoft's AI lab.
How it went
Microsoft says AutoX now monitors 100% of contact center calls for compliance, up from a manual sample of 5 to 10%.
Worth knowing
The source is a vendor news release with one headline number. It does not say how flagged calls are handled or how accurate the AI reviews are.
Try it with Custom agent
Here are this week's call transcripts from <call folder> and our compliance checklist <checklist file>. For each call, check every checklist item, mark pass or fail with the exact line that shows it, and give an overall result. Then list the calls that failed, grouped by agent and by rule, so a supervisor can review them.
Swap the parts in angle brackets for your own details. This prompt is adapted from the case, not copied from the original.
What this label means: Vendor case study
The maker describes it on its own site or launch post. It shows what the product is meant to do, with no independent result.
Summarised from the original source; the link has the full story. The people and companies here are not Wize customers.