Extracting contract insights with PwC’s AI-driven annotation on AWS

This article describes PwC's AI-driven annotation (AIDA) solution built on AWS for extracting structured insights from contracts using LLMs and automated workflows.

• AIDA reduces manual contract review time by up to 90% through AI-powered extraction
• Three core capabilities: template-based extraction, document-level chat, global cross-document chat
• Uses Amazon Bedrock LLMs with Retrieval Augmented Generation for accurate, cited responses
• Processes contracts asynchronously via Amazon ECS/Fargate with Amazon SQS coordination
• Stores documents in Amazon S3, structured data in Amazon RDS with encryption at rest
• Integrates with downstream systems via Lambda, EventBridge, and SQS for workflow automation
• Includes human-in-the-loop review queue before forwarding extracted insights downstream
• Provides operational dashboards via Amazon QuickSight for performance monitoring

AIDA transforms unstructured contracts into searchable, structured insights, enabling faster legal and compliance workflows while maintaining security and traceability across enterprise systems.