Sun Finance automates ID extraction and fraud detection with generative AI on AWS
91% cost cut. Sun Finance automated ID verification in under 5 seconds using AWS Bedrock—and accuracy jumped from 79.7% to 90.8%.
Why it matters
This is a real-world deployment case study showing how enterprises are combining specialized OCR with LLMs to solve compliance-heavy workflows. The dramatic cost and speed gains (91% per-doc savings, 20h → 5s) demonstrate AI's ROI in production fintech systems—a playbook other regulated industries will chase.
The key facts
6 to knowExtraction accuracy improved from 79.7% to 90.8%
Per-document processing cost cut by 91%
Processing time reduced from up to 20 hours to under 5 seconds
Stack: Amazon Bedrock, Amazon Textract, Amazon Rekognition
Architecture: serverless fraud detection with vector similarity search
Hybrid approach (specialized OCR + LLM structuring) outperformed single-tool solutions
Go to the source
AWS Machine Learning Blogaws.amazon.com
Publisher excerpt: In this post, we show how Sun Finance used Amazon Bedrock, Amazon Textract, and Amazon Rekognition to build an AI-powered identity verification (IDV) pipeline. The solution improved extraction accuracy from 79.7% to 90.8%, cut per-document costs by 91%, and reduced processing time from up to 20…