What challenges did Tungsten Automation face before AI implementation?

Insurance brokers and carriers manage vast volumes of complex paperwork and regulatory requirements. Manual document processing is slow, error-prone, and hinders customer responsiveness. Comparative review of policy documents is tedious and time-consuming. Rising compliance demands and risk of errors threaten service quality.

What was the impact of the AI solution?

Dramatic improvement in claims and policy document review speed and accuracy. Reduced manual labor, fewer errors, and improved compliance. Faster and more accurate responses to customer queries and comparative analysis. Staff freed to work on higher-value tasks and customer engagement.

Source

MicrosoftExpanded

Insurance Copilot accelerates claims and policy processing for insurers

Use case typeClaims automationUpdated Jun 13, 2026

Tungsten Automation, in partnership with ImageTech Systems and BlueOrange Digital, developed an insurance-specific Copilot solution leveraging Azure-integrated workflow automation, advanced OCR (TotalAgility platform), and large language models. This Copilot automates and streamlines the review and comparative analysis of complex insurance claims and policy documents. Key information is extracted using OCR, ingested into LLMs for training and querying. The solution uses a multi-agent architecture to handle multi-step workflows, improving accuracy compared to simple one-step automation. Benefits include faster claims/policy review, reduced human error, and allowing staff to focus on higher-value tasks.

Organization: Tungsten Automation
Industry: Insurance
Location: France
Published: August 2024

Reported outcomes

Strategic outcomes

New product / capabilityAutomated claims and policy document reviewRisk & complianceImproved compliance and reduced errorsCustomer experience & trustFaster customer query responsesEmployee experienceFreed staff for higher-value work

Primary read

Use case focus

Showing 2 of 2

1AI Copilot for Insurance Claims and Policy Review
2Multi-Agent Document Extraction and Comparison

Insurance brokers and carriers manage vast volumes of complex paperwork and regulatory requirements.
Manual document processing is slow, error-prone, and hinders customer responsiveness.
Comparative review of policy documents is tedious and time-consuming.
Rising compliance demands and risk of errors threaten service quality.

Built Copilot using Tungsten’s TotalAgility OCR integrated with Microsoft Azure for data ingestion and workflow automation.
Data extraction and document details passed into BlueOrange Digital-trained LLMs for advanced query answering and comparison.
Multi-agent architecture enables multi-step workflow execution for higher accuracy.
Integrated into insurers’ existing systems to support workflow automation and compliance.

Technologies

Azure TotalAgility LLM Copilot

Dramatic improvement in claims and policy document review speed and accuracy.
Reduced manual labor, fewer errors, and improved compliance.
Faster and more accurate responses to customer queries and comparative analysis.
Staff freed to work on higher-value tasks and customer engagement.

Architecture

Insurance Copilot solution integrates Tungsten’s TotalAgility OCR (for data ingestion) via Microsoft Azure into LLM-powered multi-agent workflows built by BlueOrange Digital. OCR extracts document data, which LLMs process for querying/comparison; multi-step agentic architecture drives workflow automation. Integrated into insurers’ business systems for seamless handoff of information and compliance management.

Sources & evidence1

ExpandedExpanded

The same organization appears in newer AI deployment evidence.