What challenges did Alnylam Pharmaceuticals face before AI implementation?

Manual intake, analysis, categorization, and investigation of hundreds of drug product quality complaint reports took 3–4 days. Information searching previously took about 15 minutes.

What was the impact of the AI solution?

Reduced information searching time from about 15 minutes to about 30 seconds. AskALNY is used by more than 2,000 employees and 1,000 contractors.

Source

Evidence: Medium50/100

Alnylam Transforms Product Complaint Management with Generative AI Using Amazon Bedrock

Use case typeDocument automationUpdated Jun 13, 2026

Alnylam Pharmaceuticals uses generative AI to streamline product complaint management and internal knowledge access. The company built a product complaints intake and triage prototype with Amazon Bedrock and Amazon S3, and a Slack-integrated internal assistant called AskALNY for employee information retrieval.

Organization: Alnylam Pharmaceuticals
Industry: Pharma
Location: United States
Published: May 2026

Reported outcomes

Time: 3–4 days

Time & speed

Time: Approximately 15 minutesTime: Approximately 30 seconds

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

Normalized claim

Time: 3-4 days decrease

AWS Customer StoriesMay 13, 2026Customer storyInferred claimMedium evidence strength

Reduced product complaint triage time from 3–4 days to hours.

Normalized claim

Time: 15 minutes decrease

AWS Customer StoriesMay 13, 2026Customer storyInferred claimMedium evidence strength

Reduced information searching time from about 15 minutes to about 30 seconds.

Normalized claim

Time: 30 seconds decrease

AWS Customer StoriesMay 13, 2026Customer storyInferred claimMedium evidence strength

Reduced information searching time from about 15 minutes to about 30 seconds.

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

Customer: Alnylam Pharmaceuticals
Provider: AWS
Maturity: Unknown
Linked source: AWS Customer Stories

No explicit deployment-stage evidence found.

Customer identity supportedSource describes one deploymentMaturity evidence evaluated

Primary read

Use case focus

Showing 3 of 4

1Workflow Automation
2Document Processing
3Knowledge Management

Alnylam and AWS built a generative-AI intake and triage prototype in about 3 months using Amazon Bedrock to interpret complaint content, generate decision support output, and map data into the quality system.
The solution stores extracted complaint data in Amazon S3, uses confidence scoring and exception handling, and automates record creation in the quality system.
Alnylam also built AskALNY, a Slack-integrated internal assistant using Amazon Bedrock and retrieval-augmented generation to help employees find information and summarize documents.

Technologies

Amazon Bedrock Amazon Simple Storage Service Amazon Q Business Retrieval-Augmented Generation Slack

Reduced information searching time from about 15 minutes to about 30 seconds.
AskALNY is used by more than 2,000 employees and 1,000 contractors.

Architecture

The solution uses Amazon Bedrock foundation models to read complaint content, infer and generate outputs, apply rule-based categorization, and create records automatically. Extracted complaint data is stored in Amazon S3. The system adds confidence scoring across comprehensiveness, accuracy, relevance, and clarity, and routes exceptions for human review. A separate internal assistant, AskALNY, is integrated with Slack and uses retrieval-augmented generation for more accurate employee responses.

Sources & evidence1

Evidence: Medium50/100Evidence strength