Novo Nordisk accelerates clinical documentation and drug development with Claude

Novo Nordisk faced severe bottlenecks in regulatory documentation, with clinical study reports (CSRs) reaching 300 pages and taking months to produce manually; staff averaged only 2.3 CSRs annually, and each day of delay in bringing medicines to market could cost up to $15 million in potential revenue. The company built NovoScribe, an AI documentation platform using Claude Code with retrieval-augmented generation on domain-expert-approved text, deployed on Amazon Bedrock and MongoDB Atlas. Clinical documentation time fell from 10+ weeks to 10 minutes, device verification protocol resource needs dropped 95%, CSR writing time was cut by 90%, and complete study booklets and patient guides are now produced in under a minute, all while keeping a small development team of 11. Novo Nordisk received positive regulatory feedback and plans to expand Claude use across medical affairs, including automating Common Technical Documents (CTDs) for drug approval submissions.

Anthropic builds the Claude model family, including a dedicated Claude for Life Sciences push used across drug discovery, regulatory writing and clinical documentation.

This sits in the Build stage of the AI use-case lifecycle: Named examples from the foundation-model labs and platforms pharma teams use to actually construct, train and generate the model or application.