How FDA’s New AI System ELSA Is Revolutionizing Regulatory Review

The U.S. Food and Drug Administration (FDA) has entered a new chapter in regulatory science with the introduction of ELSA—its generative AI tool designed to enhance how the agency reviews, interprets and processes regulatory information. As technology continues to reshape industries, even the most traditional and highly scrutinized government bodies are beginning to adopt AI in meaningful ways. But what does this evolution mean for the companies that must interact with the FDA—and more importantly, for those seeking approval in a system that is still very much grounded in law and policy?

While ELSA is a government-built tool with significant implications, it’s not a one-size-fits-all solution. In this article, we’ll explore what ELSA is, how it works, and why it matters—especially for businesses trying to navigate FDA regulations with clarity and confidence.

What Is ELSA?

ELSA (an acronym, though not officially defined) is the FDA’s proprietary generative AI tool launched in 2025. It was built to support internal FDA operations, particularly by assisting staff in reviewing documents, summarizing clinical trial protocols, and comparing labeling information. Unlike publicly available AI tools, ELSA is restricted to FDA use and is trained on regulatory data rather than open-source internet content.

The goal? Increase speed, consistency, and accuracy in daily agency workflows—without compromising the scientific rigor that underpins FDA oversight.

The Technology Behind ELSA

ELSA operates on a closed, controlled AI model tailored to FDA documentation libraries. This means it’s designed to reference actual source material rather than generate speculative or unverified text. When used as intended, ELSA cites its sources directly and supports reviewers by surfacing relevant regulatory information quickly.

Importantly, ELSA does not have access to the open internet. This design choice limits its potential exposure to unverified information and mitigates the risk of leaking sensitive data. The AI is trained only up to a certain date and relies entirely on internal FDA systems, ensuring data integrity but also posing limitations for real-time or rapidly evolving topics.

Benefits for the Industry and Public Health

From an operational standpoint, ELSA offers clear advantages:

• Faster document review times
• More consistent interpretations of complex regulatory language
• Reduced human fatigue in repetitive review tasks

For public health, these benefits translate into faster evaluations of new treatments and devices, potentially accelerating access to innovation. For industry stakeholders, the possibility of faster review cycles is appealing—but it also introduces new variables in the approval process.

Challenges and Limitations

Despite its promise, ELSA is not a fully autonomous regulatory system. Its effectiveness depends on human reviewers, institutional oversight, and careful implementation. Transparency is a primary concern: will drug sponsors or manufacturers know when AI influences a regulatory decision?

Another challenge is version control and training limitations. Because ELSA isn’t connected to real-time updates, there may be gaps between current scientific developments and what the AI can access. For businesses relying on the most up-to-date research, this is a potential disconnect that cannot be ignored.

How Reliable Is ELSA in Making Enforcement Decisions?

It’s important to understand that ELSA is not authorized to make enforcement decisions on its own. Instead, it assists FDA staff by summarizing and comparing documents. All enforcement actions are still made by human regulators, who review, validate, and interpret the AI’s output. The tool can enhance efficiency but does not have decision-making authority in legal or compliance matters.

Will ELSA Replace Human FDA Reviewers in Enforcement Actions?

No. While ELSA can streamline internal operations, it is not designed to—and cannot—replace FDA reviewers. Enforcement decisions, whether related to warning letters, recalls, or import holds, require professional judgment grounded in law, precedent, and context. AI can aid that process, but it does not supplant it.

Can ELSA Make Errors or “Hallucinate” When Reviewing FDA Cases?

Like any AI, ELSA can make errors—particularly if used outside of its intended environment. When operating within document libraries that force citation, ELSA is designed to minimize hallucinations. However, if used incorrectly or without verification, it could produce misleading summaries or miss critical context.

This is a major reason why businesses should not rely on automated tools alone for interpreting FDA expectations. The agency’s own staff are trained to validate ELSA’s outputs—but industry stakeholders don’t have that same access or oversight.

What This Means for Companies Navigating FDA Approvals

For companies submitting to the FDA, ELSA represents both opportunity and complexity. Faster reviews are possible—but so are shifts in how materials are interpreted. That’s why legal guidance remains essential. Businesses must continue to provide documentation that meets FDA standards while also anticipating how AI might assess or flag certain elements.

Working with a regulatory attorney helps bridge that gap. It ensures your submissions are not just complete—but also strategic, defensible, and aligned with current agency workflows.

The Future of AI in FDA Operations

ELSA is just the beginning. The FDA has publicly stated its vision of moving toward a real-time regulatory environment—where submissions, inspections, and surveillance can happen faster and with greater efficiency. Future AI tools may assist with predictive toxicology, adaptive clinical trials, and post-market surveillance.

But that vision still requires human oversight. The legal framework surrounding FDA approval has not changed—and will likely become more nuanced as AI becomes a larger part of the process.

Trusted Guidance in a Changing Landscape

As the FDA integrates tools like ELSA into its operations, companies face a shifting regulatory environment where both speed and scrutiny are increasing. Having a legal partner who understands this evolution is key.

Marc Sanchez, a regulatory attorney, brings deep experience in FDA law, helping clients navigate approvals, compliance, and enforcement with clarity and confidence. While AI like ELSA supports the FDA’s work, it does not replace the need for strategic legal interpretation. For companies operating at the intersection of innovation and regulation, that distinction is more important than ever.Need help navigating FDA approvals in the age of AI? FDA ATTY provides expert legal guidance tailored to your needs. Contact us today to stay ahead in this evolving regulatory landscape.