FDA Launches AI Tool That Could Revolutionize Food Safety

The way you shop for food might be quietly changing, as regulators bring AI into the process behind the scenes. 

The U.S. Food and Drug Administration (FDA) has introduced Elsa, a new internal AI tool designed to assist agency staff in reviewing safety data, comparing product labels, and streamlining regulatory tasks. Elsa won’t write labels or communicate directly with consumers, but its behind-the-scenes role could influence how food safety risks are flagged, how inspections are prioritized, and how quickly the FDA responds to issues.

That matters because AI is already becoming more common in the food system, from formulation and ingredient swaps to supply chain modeling. While Elsa itself isn’t involved in food product development or label writing, its debut comes at a time when lawmakers and consumer advocates are asking a related question: Should food labels reflect when AI played a role in creating what’s inside?

There’s no federal law requiring that kind of disclosure — at least, not yet. 

But the conversation is gaining momentum. A December 2024 survey from Ingredient Communications, a market research and consulting firm in the food and beverage industry, found that 83% of consumers want companies to disclose when AI is used in the development or production of food. Nearly two-thirds said they don’t think AI-assisted products should be labeled “natural,” reflecting a broader discomfort with invisible tech shaping what ends up on store shelves.

“Many food and beverage companies have rushed to embrace the benefits of AI technology, but it’s important they take care to consider how consumers feel about this,” said Richard Clarke, managing director of Ingredient Communications. “To avoid a backlash, manufacturers should reflect on whether they’re being sufficiently transparent.”

Elsa isn’t part of that labeling decision, but it’s already part of the system that could support it. Built in a secure government cloud, the tool helps FDA staff move more quickly through complex safety reports, labeling documents, and inspection planning. The agency says Elsa is already used to summarize adverse event reports, identify labeling inconsistencies, and flag issues that may require closer scrutiny.

For shoppers, Elsa’s behind-the-scenes role could impact how quickly important information is shared when something goes wrong. Serious food recalls can take several weeks to be officially classified by the FDA, meaning alerts might not reach the public until well after a problem is identified. Food safety professionals have noted that in some cases, it can take three to five weeks or longer for a recall to go through the system and reach the public.

By helping FDA staff scan safety reports and identify high-risk trends more quickly, Elsa may  shorten that timeline. While it won’t eliminate delays entirely, the tool could support faster responses during outbreaks or contamination events, improving how and when consumers are notified.

The agency has acknowledged that Elsa is still in its early stages of rollout. Some FDA staff have flagged concerns about Elsa’s accuracy with large data sets and the need for human oversight, as reported by BioSpace, STAT News, and NBC News. However, Elsa’s deployment represents a shift: AI is now part of how food safety decisions are made, even if it’s not yet reflected on the package.

For now, Elsa remains behind the scenes. But the questions it raises — about visibility, trust, and the future of food labeling — are starting to take shape in plain sight.