FDA Unveils Gen AI Tool 'Elsa' to Boost Internal Efficiency Across the Agency

Story BySaloni Tyagi

•

1 month ago

FDA Unveils Gen AI Tool 'Elsa' to Boost Internal Efficiency Across the Agency — Source: X

Elsa assists with reading, writing, and summarizing tasks. It can summarize adverse events to support drug safety profile assessments and rapidly compare packaging inserts.

The US Food and Drug Administration (FDA) has launched Elsa, a generative AI tool to enhance staff efficiency across departments, from scientific reviewers to investigators.

Elsa is part of a broader modernization effort to leverage AI to improve internal workflows and better serve the public.

“Following a very successful pilot program with FDA’s scientific reviewers, I set an aggressive timeline to scale AI agency-wide by June 30,” said FDA Commissioner Marty Makary.

“Today’s rollout of Elsa is ahead of schedule and under budget, thanks to the collaboration of our in-house experts across the centers,” he added.

Developed within a high-security GovCloud setup, Elsa ensures internal document access remains confidential, with no training on industry-submitted data. This measure helps maintain the integrity of the sensitive information managed by the agency.

“Today marks the dawn of the AI era at the FDA with the release of Elsa, AI is no longer a distant promise but a dynamic force enhancing and optimizing the performance and potential of every employee,” said FDA Chief AI Officer Jeremy Walsh.

“As we learn how employees are using the tool, our development team will be able to add capabilities and grow with the needs of employees and the agency,” Jeremy added.

Once the FDA receives an application for potential drug approval, it has six to ten months to decide.

Elsa assists with reading, writing, and summarizing tasks. It can summarize adverse events to support drug safety profile assessments and rapidly compare packaging inserts.

"Elsa offers a secure platform for FDA employees to access internal documents while ensuring all information remains within the agency. The models do not train on data submitted by regulated industry, safeguarding the sensitive research and data handled by FDA staff," the FDA said.

Currently, Elsa is being used to accelerate clinical protocol reviews, assist in scientific evaluations, and prioritize inspections. With its large language model foundation, Elsa supports reading, writing, summarizing, and coding tasks. This includes summarizing adverse events, performing label comparisons, and aiding database development for nonclinical applications.

This rollout represents the FDA’s first significant step in its AI integration journey. Plans involve expanding Elsa’s role in data processing and other generative AI functions to support the agency’s mission.