What “Generally Recognized as Safe” Really Means
A new FDA proposal would close a large information gap around food ingredients, but mandatory notification would still not mean conventional premarket approval.
In short
What happened. The US Food and Drug Administration has proposed requiring companies to notify it when they rely on a conclusion that a substance is generally recognized as safe, or GRAS, for a particular food use.
What it means. Today, a company can reach some GRAS conclusions without informing FDA. The proposal would close much of that information gap, but a notice would still not be the same thing as conventional premarket approval.
Risks and impact. The present gap can leave regulators and consumers unaware of some uses and their supporting evidence. The opposite mistake is to treat every GRAS ingredient as untested or unsafe.
What can be done. Read an ingredient claim precisely: check the substance, its intended use, whether a public FDA notice exists and what evidence supports that use.
What to watch. Public comments close on December 9, 2026. The final text, timing and treatment of substances already on the market will determine how much changes in practice.
Shown as a summary because of your reading settings.
What happened
The phrase looks reassuring enough to print on a small green badge: “generally recognized as safe.” Yet it is a legal category, not a gold star placed on an ingredient after one standard FDA approval process.
Under US food law, a substance intentionally added to food is generally treated as a food additive requiring premarket review and approval unless an exception applies. One exception covers a use that qualified experts generally recognize as safe. That recognition must concern the substance under its intended conditions of use—not the substance in every quantity, recipe or population.
Scientific GRAS conclusions require the same quantity and quality of safety evidence required for food-additive approval, FDA says. The route is different. A GRAS use is excluded from the food-additive approval requirement, and the current notification programme is voluntary. A company can submit its conclusion and evidence; FDA may reply that it has no questions at that time. A company can also reach an independent conclusion without notifying the agency.
On August 10, FDA announced a proposed rule. Published in the Federal Register on August 11, it would convert covered GRAS notifications from voluntary to mandatory. Comments are due December 9. It is a proposal, not the law now in force.
What the evidence supports
The strongest evidence here is the proposed rule itself. It says firms would still be able to reach their own GRAS conclusions, but would have to submit covered conclusions to FDA. A notice would include the intended use and the data supporting the conclusion. FDA says that information would help it identify uses that may instead require food-additive review.
The proposal also contains limited exceptions. For some substances already in interstate commerce before a final rule takes effect, a time-limited streamlined submission could place core information on a public list. FDA could then use post-market work to decide whether a full notice or food-additive petition is needed.
The oversight problem is not new. In 2010, the US Government Accountability Office reported that companies could determine a substance was GRAS without FDA’s approval or knowledge. It said the agency was not systematically ensuring the continued safety of current GRAS substances and recommended stronger oversight.
What the evidence does not support is a shortcut from “self-affirmed” to “unsafe.” Nor does a submitted notice prove that no concern can ever emerge. Safety depends on exposure and intended use, and scientific knowledge can change.
How the story is being framed
The reform can be framed in three different ways. Each catches part of the story.
The public-health frame calls the current arrangement a loophole. It correctly points to a basic governance problem: an agency cannot easily prioritise substances it does not know are being used. Mandatory notices and a more complete public inventory would make hidden decisions visible. But visibility is not the same as a full premarket approval decision for every substance.
The industry frame stresses that GRAS is a demanding scientific standard, not permission to guess. FDA says a scientific GRAS conclusion requires evidence comparable in quantity and quality to food-additive approval, and the conclusion must be generally available and accepted among qualified experts. That is important. Yet a strong standard on paper does not erase questions about conflicts of interest, documentation or consistent application when the company relying on the conclusion selects the experts and may not currently notify FDA.
The consumer frame often asks a simpler question: “Has FDA approved this?” The honest answer may be no, even when a use is lawfully considered GRAS. A “no questions” letter means the agency has no questions at that time based on the submitted and available information. The proposed rule explicitly says such a letter is not an FDA affirmation that the substance is GRAS.
The evidence leans strongly toward greater transparency. It does not support turning a regulatory-process story into a verdict on every ingredient that passes through that process.
The background
Congress created the modern food-additive framework in 1958. The aim was premarket scrutiny for substances added to food, while avoiding a formal petition for familiar uses whose safety was already broadly accepted. The GRAS exception covered safety shown through scientific procedures or, for substances used in food before January 1, 1958, experience based on common use.
That balance produced two questions that sound similar but are not. First: is there enough evidence for qualified experts generally to recognise a particular use as safe? Second: did FDA itself conduct the conventional premarket approval process for that use? GRAS answers the first and, when valid, removes the use from the second.
FDA proposed a voluntary notification system in 1997 and formalised the procedure in a 2016 final rule. Notification created a public record and allowed the agency to evaluate whether the submitted basis supported the notifier’s conclusion. It did not eliminate independent conclusions.
That is why one label can hide several histories. A familiar ingredient may be listed in regulation. Another may have a public notice followed by an FDA response. A third may rest on a conclusion never submitted to the agency. The proposed rule is designed to reduce that third category and give FDA information for follow-up.
It would not freeze science. A conclusion concerns an intended use at a point in time. A new manufacturing process, higher exposure, a different population or new toxicology can change the assessment. “Safe” in food regulation never means that dose, context and later evidence have stopped mattering.
Who it touches
For a shopper, the system is almost perfectly arranged to cause confusion. The ingredient list is visible. The route by which each use entered the food supply usually is not.
A parent may see a chemical name and assume danger because it is unfamiliar. Another shopper may see “GRAS” and assume FDA performed the same review it uses for an approved food additive. Both reactions compress a long regulatory record into a feeling.
Manufacturers face a different problem. A mandatory system would require time, evidence organisation and submission. Smaller firms may feel that burden more sharply. Regulators would receive more information, but information still has to be reviewed, prioritised and acted upon.
The useful question for all three groups is not whether the ingredient sounds natural. It is whether the evidence fits the actual use and whether the decision is visible enough to be challenged when new facts arrive.
The deeper story
Food labels invite a small act of theatre. We stand in an aisle, turn a package over and try to convert twelve unfamiliar words into certainty before someone reaches around us for the cereal.
The words themselves rarely carry enough information. Vinegar and black pepper can be GRAS. So can substances with names that sound as if they belong in a laboratory, because laboratories are where chemistry receives precise names. Familiarity is not a safety test, and unfamiliarity is not evidence of harm.
A better four-line decoder works like this:
- Name the use. Safety applies to a substance under stated conditions, including amount and purpose.
- Name the route. “FDA approved,” “GRAS,” and “FDA has no questions” are not interchangeable phrases.
- Find the record. FDA maintains public GRAS notice information. No listing does not itself prove illegality, because independent conclusions are currently possible, but it reveals what the agency and public can inspect.
- Ask what changed. New exposure, manufacturing methods or research may matter more than the age of the ingredient.
This does not turn a shopper into a toxicologist. It does something more modest and more useful: it prevents a process label from pretending to be a complete answer.
The proposed rule is fundamentally about that missing middle. It would not guarantee that every scientific judgement is correct. It would make far more of those judgements visible—to regulators, researchers, competitors and citizens who can ask better questions.
A window is not the same as a lock. But when a public system has operated with a dark room, adding a window is a serious change.
Something to sit with
When a label says “safe,” do you instinctively hear a scientific conclusion, a government approval or a marketing promise?
How much confidence should transparency add when the underlying judgement can still be disputed?
Which matters more to trust: eliminating every uncertainty, or making the evidence and decision process open to challenge?
Sources
- Federal Register / US Food and Drug Administration — https://www.federalregister.gov/documents/2026/08/11/2026-16296/sub...
- US Food and Drug Administration — https://www.fda.gov/food/food-ingredients-packaging/generally-recog...
- US Government Accountability Office — https://www.gao.gov/products/gao-10-246
- Reuters — https://www.reuters.com/legal/litigation/us-health-agency-moves-inc...
We report facts from the sources above in our own words and link to the originals. Interpretation is ours, not theirs.
If the FDA's proposed GRAS rule becomes final, what would it primarily require?
The proposal would make covered GRAS notices mandatory and improve transparency, while GRAS uses would remain legally distinct from conventionally approved food additives.
♻︎ Free to republish
Copy this HTML into your CMS. Credit line and licence are included. Republish our work — free
Every headline has a deeper story. This is ours.
What we are doing here