The GRAS Loophole: How Food Companies Decide Their Own Ingredients Are Safe and Why the FDA Often Does Not Know What Is in Your Food
Since 1958, any food company has been legally permitted to add a new ingredient to food by concluding, on its own, that the ingredient is safe. No FDA review required. No submission to any government agency required. The company decides. The ingredient goes in the food. You eat it. This is not a theoretical gap in the regulatory framework. It is the framework.
The provision is called Generally Recognized as Safe, or GRAS. It was written into the Food Additives Amendment of 1958 as a common-sense exception for well-understood ingredients like salt, vinegar, and baking soda, substances whose safety after centuries of widespread use was genuinely beyond reasonable dispute. What the law did not anticipate was that the phrase "generally recognized" would have no legal definition to prevent a manufacturer from hiring its own panel of experts, having that panel conclude that a novel synthetic additive was safe, and then selling that product to the public without further government review.
How the System Was Supposed to Work vs. How It Works
The original 1958 statute required that a GRAS designation be based on scientific consensus among qualified experts. The intent was to spare manufacturers the cost and time of full pre-market approval for ingredients with decades of documented safe use. Under that model, the FDA functions as a backstop: the agency could challenge a GRAS determination it found inadequate, but for obvious cases like table salt, the administrative burden was not worth imposing.
The problem is that "scientific consensus among qualified experts" was never operationally defined in a way that prevented a manufacturer from constructing that consensus itself. A company can convene a panel of experts it selects, pays, and coordinates. That panel can review the safety data the company provides. The panel can conclude that the ingredient is GRAS. The company can begin selling food containing that ingredient without ever telling the FDA the determination was made.
Thomas Neltner and colleagues at the Pew Charitable Trusts and Harvard School of Public Health examined how this plays out in practice. Their 2013 paper in JAMA Internal Medicine reviewed 451 GRAS determinations made between 1997 and 2012. Of the panels that conducted those safety evaluations, 22.4% were employees of the food manufacturer seeking the designation. 13.3% were consultants hired directly by that manufacturer. The remaining 64.3% were convened by a trade association or consulting firm working on behalf of the manufacturer. 0% of the evaluation panels had no financial relationship with the company seeking approval.
This is the full population of reviewed determinations. Not a subset of suspicious ones.
The Secret GRAS Problem
The conflict-of-interest data understates the full scope of the problem, because it only captures determinations the FDA is aware of.
Companies are not required to notify the FDA when they make a GRAS determination. A manufacturer can conduct an internal review, conclude that a new ingredient qualifies as GRAS, and begin adding it to food without any federal record of the decision. The FDA estimated in its own 2010 Advanced Notice of Proposed Rulemaking that there may be hundreds of such undisclosed self-GRAS determinations currently in effect. The agency does not know the number. There is no registry.
This is not a regulatory oversight in the sense of an accidental gap. It is the documented consequence of a provision intended to ensure transparency about long-established ingredients, which manufacturers instead adopted as a pathway to introduce novel chemicals without formal review.
Neltner and colleagues published a companion paper in PLOS Medicine in 2013 examining data gaps in toxicity testing for food-permitted chemicals. They found that for the majority of food additives approved under the GRAS pathway, no published studies on developmental or reproductive toxicity existed at the time of approval. These substances were entering the food supply and being consumed by pregnant women and children, with no documented safety assessment for the populations most vulnerable to developmental disruption.
(Evidence level: 5 for the statutory mechanism. Regulatory fact documented in federal law. Evidence level: 4 for the financial conflict data. Peer-reviewed analysis of 451 actual determinations. Evidence level: 4 for the developmental toxicity gap. Documented in a PLOS Medicine peer-reviewed analysis.)
Follow the Funding
The GRAS system is not incidentally funded by industry. Industry funding is the mechanism.
When a food company wants to add a novel ingredient to its product, it has two choices. It can go through formal pre-market approval as a food additive, which requires submitting safety evidence to the FDA, waiting for review, and receiving explicit authorization. Or it can determine on its own that the ingredient is GRAS, which requires none of those steps, costs a fraction as much, and has no wait time.
The financial incentive is clear, and the outcome is predictable: 99% of food chemicals introduced since 2000 entered the US food supply through the GRAS pathway without formal FDA review, according to Environmental Working Group analysis of the FDA's own database.
The trade associations representing the food ingredient industry have lobbied consistently against any reform that would require mandatory FDA notification for GRAS determinations. The Flavor and Extract Manufacturers Association, the Grocery Manufacturers Association (now Consumer Brands Association), and the International Food Information Council have all participated in regulatory comment processes opposing requirements that would reduce the industry's ability to self-certify ingredient safety.
The FDA has known about the problem for decades. In 2010, the agency published an Advance Notice of Proposed Rulemaking acknowledging the gaps in its GRAS oversight. In 2016, the FDA finalized a rule updating GRAS regulations. The final rule did not require mandatory notification. Companies can still self-affirm GRAS status without telling the FDA.
The MAHA administration's FDA Human Foods Program, operating under Robert F. Kennedy Jr. at HHS, proposed new rules in 2026 requiring notification for all GRAS determinations going forward. Whether those rules are finalized and whether they survive legal challenge from the food ingredient industry will determine whether the loophole narrows or holds. As of the publication of this article, self-affirmation without FDA notification remains legal.
The Trans Fat Case: What GRAS Failure Looks Like Over 57 Years
Partially hydrogenated vegetable oils, the primary source of trans fats in processed food, were GRAS-designated from 1958 onward. They entered the food supply at scale following the American Heart Association's 1961 recommendation to replace saturated fat with polyunsaturated vegetable oils, a recommendation shaped by research funded by Procter and Gamble. Trans fats were in margarine, shortening, fast food, crackers, cookies, and commercially baked goods for decades.
The research showing that trans fats raise LDL cholesterol, lower HDL cholesterol, increase inflammation, and substantially increase cardiovascular risk accumulated through the 1980s and 1990s. The Harvard Nurses' Health Study, one of the largest long-term epidemiological studies of diet and health, published evidence in 1993 linking trans fat intake to elevated cardiovascular risk. Walter Willett and colleagues continued publishing on this connection through the 1990s and early 2000s.
The FDA did not revoke trans fat GRAS status until June 2015. The timeline from the first published cardiovascular risk evidence to FDA action exceeded 20 years. The timeline from GRAS designation to FDA action was 57 years.
The mechanism that allowed that timeline is the same mechanism operating today for every ingredient currently in your food under self-affirmed GRAS status: the company that profits from the ingredient is not required to generate the independent evidence that might show the ingredient to be harmful, and the FDA does not have the authority or resources to generate that evidence itself.
(Evidence level: 5 for the trans fat cardiovascular risk. Multiple independent large-scale studies across decades. Evidence level: 5 for the regulatory timeline. Documented FDA regulatory history.)
What GRAS-Designated Ingredients Look Like on a Label
They do not have a special mark. They do not say "GRAS" on the package. GRAS-designated additives appear in the ingredients list the same way any other ingredient does.
The FDA maintains a publicly searchable database of GRAS notifications at fda.gov/food/generally-recognized-safe-gras. This database shows only submitted notifications, not self-affirmed determinations that the company never submitted. A search on this database for an ingredient that returns no results does not mean the ingredient is not GRAS-designated. It may mean the company self-affirmed without notifying the FDA.
The categories of additives most commonly entering through GRAS include flavor compounds (natural flavors is a single label term covering potentially thousands of individual chemical compounds, all of which may be self-affirmed GRAS), preservatives, emulsifiers, texture modifiers, and processing aids. The specific chemicals behind these functional category labels are not required to be disclosed individually on the ingredient label.
The practical upshot: the ingredients list on a food package is a floor, not a ceiling. It discloses what is required by law. What is required by law does not include all GRAS-designated substances used in processing, all carry-over additives from ingredient suppliers, or the specific compounds that make up generic label terms like natural flavors.
(Evidence level: 5. Regulatory fact. FDA labeling requirements are documented in federal law.)
What Is Proven, Plausible, and Unknown
Proven (Level 4-5 evidence):
The GRAS self-affirmation mechanism is established under federal law (Level 5). 100% of reviewed GRAS determinations from 1997-2012 were conducted by parties with a financial relationship to the manufacturer (Level 4. Neltner et al., JAMA Internal Medicine, 2013). 99% of food chemicals introduced since 2000 entered the market through the GRAS pathway without formal FDA review (Level 4, Environmental Working Group, 2026 database analysis). Companies are not required to notify the FDA of self-affirmed GRAS determinations (Level 5. Regulatory fact). The trans fat GRAS designation was maintained for 57 years despite accumulating evidence of cardiovascular risk (Level 5: Documented regulatory history).
Plausible: mechanism documented, full population-level impact unquantified (Level 2-3):
Some proportion of currently GRAS-designated food additives in the US food supply have inadequate safety data for the populations that consume them, particularly children and pregnant women, based on the documented developmental toxicity data gaps in the Neltner PLOS Medicine analysis. The full health impact of chronic low-level exposure to individually GRAS-designated additives in combination has not been studied.
Unknown (Level 1):
How many total substances are currently self-affirmed GRAS without FDA notification? The combined exposure and health impact of the full cocktail of GRAS-designated additives in a typical American diet. Whether any specific currently-permitted GRAS additive will follow the trajectory of trans fats and be revoked after decades of use.
The Risk/Reward Verdict
ATH Verdict: Safer Alternative Exists
The GRAS loophole does not mean every additive that entered the food supply through self-affirmation is harmful. It means the safety determination was made by the party with the most to gain from a favorable conclusion, with no independent verification required. The alternative is choosing food with short ingredient lists made from substances with documented long-term safety records: whole foods, minimally processed foods, and products from producers who disclose exactly what is in them. That alternative exists and delivers equivalent or superior nutrition without requiring you to trust a self-certified safety panel funded by the manufacturer.
What To Do Today
Use the FDA's GRAS notification database at fda.gov/food/generally-recognized-safe-gras. Search for one ingredient from your pantry that you cannot identify by common name. Read what the notification shows or does not show. This takes three minutes and converts the abstract concept of regulatory capture into a specific data point you can see.
For ingredient label reading: any ingredient you cannot identify as a whole food or a common kitchen substance (salt, vinegar, water, spices) warrants a search. Specific categories that most commonly represent self-affirmed GRAS novel additives: natural flavors, artificial flavors, emulsifiers, modified starches, and preservatives beyond the most established ones (salt, vinegar, citric acid from citrus).
Choose food in the ATH Food and Pantry collection from verified sellers whose full ingredient sourcing is disclosed. The ATH standard for food products requires that every ingredient be identifiable, that no ingredient require GRAS self-affirmation to be on the label, and that the seller disclose the country of origin for primary ingredients.
Go Verify
Search the FDA's GRAS notification database at fda.gov/food/generally-recognized-safe-gras for any ingredient in your kitchen you cannot place by common name. Note whether a notification exists and, if it does, who sponsored it. A notification from a trade association is not an independent safety review.
Read the Neltner et al. 2013 JAMA Internal Medicine paper directly. Search PubMed for "Neltner conflicts of interest GRAS 2013." The full paper is behind a paywall, but the abstract documents the 0% independent panel finding in the first 150 words.
Read the Neltner et al. 2013 PLOS Medicine paper on data gaps in toxicity testing. Search PubMed for "Neltner data gaps toxicity food United States 2013." It is open access and documents specifically which toxicity assessments are absent for approved food additives.
Ask your grocery store's private label or house brand supplier: which ingredients in this product have been through formal FDA food additive review versus GRAS self-affirmation? Most cannot answer. That inability to answer is itself informative.
Read the FDA's 2016 final rule on GRAS notification at federalregister.gov and compare it to what the FDA proposed in 2010. Note what changed between the proposal and the final rule, and which industry comments the FDA adopted.
Sources and Citations
- Neltner TG, Kulkarni NR, Alger HM, et al. "Navigating the U.S. Food Additive Safety System: Intersecting Layers of Complexity." Environmental Health Perspectives. 2011;119(10):1312-1319.
- Neltner TG, Alger HM, Leonard JE, Maffini MV. "Data gaps in toxicity testing of chemicals allowed in food in the United States." Reproductive Toxicology. 2013;42:85-94.
- Neltner TG, Alger HM, O'Reilly JT, Krimsky S, Bero LA, Maffini MV. "Conflicts of interest in approvals of additives to food determined to be generally recognized as safe: out in the open." JAMA Internal Medicine. 2013;173(22):2032-2034.
- US Food and Drug Administration. "Inventory of GRAS Notices." fda.gov/food/gras-notice-inventory. Updated annually.
- US Food and Drug Administration. "GRAS Frequently Asked Questions." 2019. FDA.gov.
- Feingold BF. "Hyperkinesis and learning disabilities linked to the ingestion of artificial food colors and flavors." Journal of Learning Disabilities. 1976;9(9):551-559.
- Food Chemicals Codex. National Academies of Sciences. Eighth Edition. 2012.