Novo Nordisk, a global healthcare leader focused on treating chronic conditions like diabetes and obesity, wants to get treatments to patients as quickly as possible. However, the company’s traditional method for creating clinical study reports was slow, delaying how quickly new therapies could reach regulators and patients. To accelerate the documentation process, Novo Nordisk worked alongside AWS Partner MongoDB and AWS Partner Anthropic to develop a secure, compliant solution. Now, Novo Nordisk has revolutionized its drug development process, shortening documentation timelines and dramatically reducing overhead—making it possible to bring innovative medicines to patients faster than ever before.

Opportunity | Reaching out for Expert Guidance to Accelerate Clinical Trial Reviews

Time is both a constraint and a competitive edge in pharmaceutical research. For Novo Nordisk, generating clinical study reports was historically slow and labor-intensive, taking up to 15 weeks and requiring the coordination of 40–50 professionals. These reports are necessary for regulatory submissions and must meet strict standards for accuracy, consistency, and traceability—factors that can make automation challenging.

As the scale and complexity of documentation increased, so did the need to rethink workflows. Novo Nordisk quickly recognized that generative AI could offer a breakthrough. The company needed external expertise to guide the development of a solution that could meet both its operational goals and its regulatory obligations. That realization prompted the company to reach out to partners with proven capabilities in implementing large language models, intelligent data storage, and cloud security. “We knew early on that this wasn’t something we could solve internally,” says Louise Skov, head of content digitalization at Novo Nordisk. “We needed expert knowledge from outside the pharmaceutical industry to move forward with confidence.”

Solution | Building a Secure, Compliant Generative AI Solution

The custom generative AI solution uses Anthropic’s Claude 3.5 deployed through Amazon Bedrock —the easiest way to build and scale generative AI applications with foundation models. The model’s precision and reliability was ideal for use in a highly regulated industry, where even minor errors in language can have serious implications. Supporting these AI operations was MongoDB, which provided a flexible and intelligent data foundation, and MongoDB Atlas, which supported near real-time collaboration. This backend is integrated with AWS Lambda so that developers can run code without thinking about servers or clusters, as well as Amazon Simple Storage Service (Amazon S3), which offers object storage built to retrieve any amount of data from anywhere. Using these services help Novo Nordisk to generate high-quality, regulator-ready documents in a fraction of the time—giving teams more time to focus on other important work.

Using these services, Novo Nordisk was able to safely test and deploy multiple generative AI models without risking data exposure or compliance gaps. “We are very certain that everything deployed within our AWS Cloud is secure, with no risk of data leakage,” says Skov. The team also created a simple-to-use interface where medical professionals can submit documents, review AI-generated summaries, and provide validation and feedback to help the system improve over time. This accuracy- and accountability-based approach keeps people involved in every step.

Outcome | Reducing Time to Generate Insights by 90 Percent to Accelerate Patient Treatment

The solution’s impact has been dramatic. What once required a team of 40–50 people working for up to 15 weeks can now be completed in just minutes by a team of three. Novo Nordisk has reduced the time to generate critical documentation by over 90 percent, freeing valuable resources and accelerating time to submission for new therapies. This leap in efficiency means better business value and greater patient benefit: Novo Nordisk can prepare regulatory submissions faster, meaning new treatments can reach patients sooner.

Through its use of generative AI, Novo Nordisk has not only solved a long-standing operational challenge but also set a new standard for what’s possible in life sciences innovation. Using AWS, Anthropic’s Claude, and MongoDB, Novo Nordisk has positioned itself as a trailblazer in the pharmaceutical industry and now can bring lifesaving treatments to patients faster than ever before.