FDA Expands Use of Advanced AI for Safety Reviews and Inspections

December 4, 2025 – The Food and Drug Administration (FDA) announced on Monday that it has deployed a version of artificial intelligence called “agentic AI” for “all agency employees” to assist with complex tasks, including safety reviews, inspections, and compliance. While the release did not specifically mention food safety, the agency is responsible for monitoring the safety of the U.S.

food supply (with the exception of meatpacking plants). And many of the activities mentioned, like pre-market reviews and inspections, apply to how the agency regulates food. “FDA’s talented reviewers have been creative and proactive in deploying AI capabilities — agentic AI will give them a powerful tool to streamline their work and help them ensure the safety and efficacy of regulated products,” FDA Chief AI Officer Jeremy Walsh said in the release.

The FDA rolled out a large language model AI called Elsa in May, which the agency says is now being used by more than 70 percent of its staff. But while large language models (LLMs) can answer questions and generate reports, agentic AI is named for its ability to “make decisions, execute actions, and optimize processes without continuous human intervention,” according to Northeastern University’s Institute for Experiential AI.

In response to questions about how the FDA specifically plans to use agentic AI in its Human Foods Program, a spokesperson emphasized the role employees will continue to play. “As always, any regulatory decisions will be made by agency experts familiar with both the state of the science and regulatory standards,” they said. According to a recent ProPublica analysis, the Human Foods Program has lost 20 percent of its staff since January, due to the Trump administration’s campaign to shrink government agencies.

Consumer Reports found the food side of the agency has not experienced the level of turmoil that the drug side has, but inspections of foreign facilities that supply the U.S. have dropped to historic lows. At the same time, the FDA is taking on new efforts tied to Health and Human Services Secretary Robert F. Kennedy’s Make America Healthy Again (MAHA) agenda.

At a virtual event hosted by the group MAHA Action on Wednesday, FDA Human Foods Program Director Kyle Diamantis pointed to a new rule that will change the agency’s process for allowing companies to designate food additives as Generally Recognized as Safe (GRAS). The FDA sent the rule, which advocates hope will increase transparency and give the agency more power to regulate food additives, to the Office of Management and Budget this week for review, which means it will likely be released for public comment soon.

(Link to this post.) The post FDA Expands Use of Advanced AI for Safety Reviews and Inspections appeared first on Civil Eats.