What challenges did Accenture face before AI implementation?

Manual authoring of Common Technical Documents for FDA submissions requires up to 100,000 hours annually for large pharmaceutical companies, causing delays and errors.

Source

ExpandedEvidence: Low35/100

Accenture Regulatory Document Authoring Solution Using AWS Generative AI Services

Use case typeCompliance automationUpdated Jun 13, 2026

Accenture developed an AI-based generative solution to automate creation of Common Technical Documents (CTDs) required for pharmaceutical regulatory submissions to the US FDA. They used Amazon SageMaker JumpStart AI21 models along with AWS Lambda, Amazon S3, Amazon SQS, AWS Step Functions, Amazon Textract, DynamoDB, Amazon Route 53, and AWS Amplify to build an end-to-end document generation and editing workflow. The solution extracts key data from testing reports, generates standardized CTDs, and provides a React web app for document upload, authentication, asynchronous processing, and browser-based document editing. By automating this labor-intensive task, the system reduces CTD authoring time by an estimated 40-45%, improves efficiency, lowers errors, and accelerates drug approval timelines.

Organization: Accenture
Industry: Pharma
Location: United States
Published: February 2024

Planned next steps

The source says the organization aims to achieve Time: −40–45%.

Why do we believe this?Outcome claims, sources, and evidence checks

Normalized claim

Time: 40-45% decrease

AWS Machine Learning BlogFeb 6, 2024Blog postInferred claimLow evidence strength

Estimated 40-45% reduction in CTD authoring time.

Why do we believe this deployment?Customer identity, provider attribution, maturity, and source checks

Customer: Accenture
Provider: AWS
Maturity: Unknown
Linked source: AWS Machine Learning Blog

No explicit deployment-stage evidence found.

Customer identity supportedSource describes one deploymentMaturity evidence evaluated

Primary read

Use case focus

Showing 3 of 3

1Document Generation
2Regulatory Compliance Automation
3Generative AI

Accenture implemented a generative AI solution on AWS that extracts data from technical reports and auto-generates CTDs in the standard format.
A React web app enables users to upload documents, authenticate securely, and access asynchronously processed results.
AWS services including Amazon SageMaker JumpStart AI21 Jurassic Jumbo Instruct and Summarize models, AWS Lambda, SQS, Step Functions, DynamoDB, Amazon Textract, and Amazon S3 power the automated workflow and document parsing.
Users can edit and review the generated documents in the browser before submission.
The system meets stringent security, control, and auditability requirements using AWS Well-Architected principles and encryption.

Technologies

Amazon SageMaker JumpStart AWS Lambda Amazon Simple Storage Service Amazon Simple Queue Service AWS Step Functions Amazon Textract DynamoDB Amazon Route 53 AWS Amplify

Automated, secure system enhanced user control and audit traceability.

Architecture

The architecture involves a React web app hosted with Route 53 DNS and Amplify authentication, Amazon S3 for document storage, SQS for decoupled job queuing, AWS Lambda functions for processing, AWS Step Functions for orchestration, Amazon Textract for document parsing, DynamoDB for data storage, SageMaker JumpStart for AI model inference and fine-tuning, and a WebSocket connection for real-time updates.

Sources & evidence1

Evidence: Low35/100Evidence strength