FDA Releases AI Tool Elsa For Clinical Review Internal Writing

The FDA has launched a generative artificial intelligence (AI) tool called Elsa

On June 2, 2025, the US Food and Drug Administration (FDA) announced on its official website.
AI tool Elsa, aimed at helping people from science critics to researchers work more efficiently.
This innovative tool aims to modernize the functions of institutions and use AI capabilities to better serve the American people.
The FDA stated that Elsa is built on a highly secure GovCloud (a cloud platform for government agencies) environment, providing FDA employees with a secure platform to access internal files while ensuring that all information is stored within the agency. These models will not be trained using data submitted by regulated industries, thus protecting the sensitive research and data processed by FDA employees.
After collaborating with scientific reviewers from the US Food and Drug Administration on a highly successful pilot project, I have set an aggressive timeline to promote artificial intelligence throughout the institution by June 30th. Thanks to the collaboration of our internal experts from various centers, Elsa’s release today was completed ahead of schedule and within budget. ”. Marty Makary, Director of the US Food and Drug Administration, said.
Jeremy Walsh, Chief Artificial Intelligence Officer of the US Food and Drug Administration, stated:
Elsa’s release marks the arrival of the era of artificial intelligence by the US Food and Drug Administration. Artificial intelligence is no longer an unattainable promise, but a powerful force that can enhance and optimize the performance and potential of every employee. As we understand how employees use this tool, our development team will be able to continuously expand its functionality and grow according to the needs of both employees and the organization.
The US Food and Drug Administration stated that Elsa’s launch is the first step in its overall journey of artificial intelligence. As the tool matures, the US Food and Drug Administration plans to integrate more artificial intelligence technologies into different processes, such as data processing and generating artificial intelligence functions, to further support the FDA’s mission.
The main features of ELSA are:
Automated document analysis – using natural language processing (NLP) to extract and analyze key data from regulatory submissions such as drug applications and clinical trial reports.
Data standardization – helps ensure compliance with structured data formats such as eCTD (Electronic Common Technical Document).
Risk identification – marking potential security issues, inconsistencies, or missing information in submitted documents.
Faster review cycle – By automating some evaluation processes, the manual workload of FDA reviewers is reduced.
Integrate with FDA system – work collaboratively with existing FDA platforms to simplify workflow.
Purpose and benefits:
Accelerate the approval process while maintaining regulatory rigor.
Improve the consistency and transparency of FDA reviews.
Reduce human errors in data extraction and analysis.
Current status:
The US Food and Drug Administration has been gradually integrating AI/ML tools such as ELSA into its regulatory processes. Although not all details are public, ELSA represents the commitment of the US Food and Drug Administration to using artificial intelligence to modernize drug and device reviews.

For more informations about Elsa. Please visit youtube video