Yseop built a generative AI regulatory document tool on Amazon Bedrock and Claude 3.5.

Medical writers spend weeks or months preparing CSRs, patient narratives and other regulator-facing documents. Yseop built a platform on Amazon Bedrock and Claude 3.5 that converts structured and unstructured data into documents, with writers reviewing the output. The AWS page says one client eliminated about 10,000 hours of writing and review time in the first year and cites four hours down to four seconds.

Yseop is a French generative AI and natural language generation vendor, used by large pharma to draft clinical study reports, patient narratives and regulatory documents

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.