Understanding the New K-beauty SPF Claim Rules: What You Need to Know
Comments on two Korean cosmetics notices closed on 18 September 2026, and the coverage has been about price: an SPF test that takes about a month and costs about six million won drops to about two days and about 3.5 million. Buried in the same package is the clause that moves more money. It narrows the waiver that lets a second, third and twelfth brand ride an already-approved sunscreen dossier without running a test at all. The test gets cheaper. The number of products that need one goes up.
What did the ministry actually put out on 18 August?
Two draft notices, numbered 2026-409 and 2026-410, went onto the Ministry of Food and Drug Safety's administrative notice board on 18 August 2026 with a comment deadline of 18 September. The first amends the Regulation on the Review of Functional Cosmetics. The second amends the Regulation on Substantiation of Cosmetic Labelling and Advertising. Both list the same desk officer in the cosmetics policy division, and both were filed as a single policy move.
Functional cosmetics is Korea's regulatory class for products that make a defined efficacy claim — sun protection, whitening, wrinkle improvement, hair-loss relief. A sunscreen does not reach a Korean shelf without either clearing review or filing a report that matches a product already reviewed. Skipping both carries up to three years' imprisonment or a fine of 30 million won (about US$21,600).
The 2026-409 draft does three things. It adds test data generated by an internationally standardised in-vitro method as an acceptable basis for setting the SPF value and the PA grade. It narrows the conditions under which a new product can skip the efficacy data package. It tightens how ingredient quantities are written into the dossier. The 2026-410 draft does one thing: it recognises in-vitro data as substantiation for an SPF or PA claim in labelling and advertising, where the notice previously recognised human application test data alone.
In-vitro here means the measurement runs on a roughened plate of polymethyl methacrylate — PMMA, a transparent plastic that stands in for skin — under a solar simulator, rather than on human volunteers whose backs are dosed and then irradiated until they redden.
Does an in-vitro result replace the human test, or sit beside it?
It sits beside it. The 2026-410 draft's stated effect is that human application test data and in-vitro test data are both accepted as substantiation. That removes the exclusivity of the human test, not the human test. Korean coverage has rendered this as measuring SPF "instead of people," which reads as substitution; the notice text is a widening of what counts. A sponsor holding an existing in-vivo result keeps it. A sponsor starting from zero now chooses.
The standards named are ISO 23675 and ISO 24443. ISO 23675 was published in December 2024 as a double-plate in-vitro method for sun protection factor. ISO 24443 determines the UVA protection factor in vitro and is the method the European recommendation on sunscreen efficacy already points to on the UVA side. A third method published the same month, ISO 23698, measures SPF by hybrid diffuse reflectance spectroscopy on human skin without inducing sunburn. Korea's draft names the first two and leaves the third out.
Why is the substantiation notice not the new duty it reads like?
Because the duty already exists. Article 14 of the Cosmetics Act obliges a marketer to be able to substantiate any factual claim it makes, and gives the ministry power to demand the evidence, with 15 days to produce it. The substantiation regulation's own requirements already accept a human application test, an in-vitro test, or research of equivalent or higher standing. Nothing in the 2026-410 draft creates an obligation to prove SPF that a Korean marketer did not carry before.
What the draft does is write the in-vitro route into the sunscreen line of that regime, so the cheaper of the two proofs is admissible at the exact point where the claim gets policed. The headline effect is not "prove it or drop the claim." It is that proof becomes affordable at the same moment a much larger set of products is told to hold some.
What did the waiver cover, and what does the draft do to it?
Article 6 of the review regulation waives the safety and efficacy package when a product matches one already reviewed on the ingredients that produce the effect — type, specification, quantity — together with dosage and directions for use. In practice the match was read on the actives. A second product built on the same approved bulk with a different fragrance, a different preservative system or a different colourant cleared the gate on the strength of someone else's data.
The draft moves the line. The additives have to match as well, with fragrance, preservatives and colourants below 1% carved out. That carve-out is the whole of the remaining flexibility, and it is narrow. A tint. A preservative system swapped to satisfy a different market's negative list. A fragrance dosed at 1.5% rather than 0.9%. Each of those now sits outside the waiver, and the product behind it carries its own efficacy data into review.
The draft bans nothing. A brand still builds on a contract manufacturer's approved base, and the contract manufacturer still sells that base to twenty customers. What moves is where the paperwork sits — from a dossier the manufacturer filed once to a test each brand commissions for its own variant.
How many products does that waiver touch?
The ministry reviewed 964 functional cosmetics in 2024 — 760 made domestically, 204 imported — against 944 in 2023. Sunscreen was the largest single function at 321 cases, about 46.9% of single-function reviews. Counting every file where UV protection appears on its own or combined with whitening, wrinkle improvement or a barrier claim, 534 of the 964 carried a UV claim. That is 55% of everything Korea reviewed that year.
Set beside that, the ministry cleared 12 new active ingredients in 2024, down from 13 in 2023 and 15 in 2022. Twelve new actives against 534 UV dossiers is the shape of the market stated in the regulator's own arithmetic: the chemistry is shared, and what varies between products is precisely the layer the draft has just pulled back inside review.
The review count is also a floor rather than a census. Korea's second route onto the shelf is the report — a filing rather than a review, available when a product matches an already-reviewed item on ingredient type, specification and quantity, on efficacy, on standards and test methods, on dosage and directions, and on dosage form. Reports carry no safety or efficacy package at all. Neither route publishes a count that maps to the number of sunscreen SKUs in Korean distribution, and the gap between 964 and that number is the space both the waiver and the report have been filling.
Is Korea leading here, or catching up?
Catching up on the method, and moving early on the second half. The European recommendation on sunscreen efficacy names ISO 24444:2019 for in-vivo SPF and ISO 24443:2021 for in-vitro UVA, and states a preference for in-vitro methods that deliver equivalent results, on the grounds that in-vivo testing raises ethical concerns. Cosmetics Europe updated its own recommendation in 2025 to introduce the two methods published in December 2024, and ISO 23675 has been taken up as a European standard. Recognition of in-vitro SPF is not a Korean first, and the drafts do not claim it as one.
What is less usual is the pairing. Most regimes that opened the in-vitro door treated it as a testing question and left the dossier rules alone. Korea has opened the door and narrowed who is allowed to walk past the test entirely, in the same package, with the same comment deadline.
What does this do to the "proven base, your label" model?
An ODM — an original design manufacturer, the contract developer that owns a formula and sells it to brands that put their own name on it — competes in Korea on development speed, regulatory clearance and fill capacity. An approved sunscreen base is among the most valuable items on that shelf, because it converts a brand's twelve-month regulatory path into a six-week one. The draft does not close that model. It re-prices it.
Part of the base's value was that it carried its own proof, and a brand differentiating on a tint or a scent inherited the SPF number attached to it. Under the draft, differentiation above the 1% line detaches the proof and the brand buys its own: about two days and about 3.5 million won (about US$2,530) on the in-vitro route, against about a month and about six million won (about US$4,330) for the human test. Won figures convert at 1,386 won to the dollar, the rate on 18 September 2026, and the time and cost estimates are the regulator's own.
For a brand launching one hero sunscreen, 3.5 million won is a rounding error against a marketing budget. For a manufacturer running forty variants off one base for forty customers, the same figure is forty tests that did not exist before, and who pays for them is a commercial negotiation that has not happened yet.
What happens to the number printed on the tube?
Korea's review regulation sets the SPF label as an integer within 20% below the measured mean, and anything at or above 50 prints as SPF50+. A product measuring a mean of 23 can print any integer from 19 to 23. That rule has always compressed a spread of real results into a single printed claim, and it is unchanged by either draft.
What changes is how many of the products carrying that claim have a measurement of their own behind it rather than an inherited one, and how quickly the ministry can ask to see it. Fifteen days, under Article 14, from a marketer whose answer until now was to point at the base.
Where does this leave the three sides of the table?
For contract manufacturers, the approved base stops being a finished asset and becomes a starting point with a test attached. Those with in-vitro capacity in house absorb the change and sell it as a service. Those who outsourced SPF testing add a line item to every variant they ship.
For brands and the buyers who stock them, the document that matters in a due-diligence pack changes. "Built on an approved base" stops being an answer to a question about SPF. The answer is the test report for that SKU, with the method named on it.
For regulators and investors reading Korea as a template, the pairing is the point. Opening a cheaper method while narrowing an exemption is neither deregulation nor tightening. It is a transfer of who holds the evidence. The cost of proof went down. The number of parties required to hold it went up. Korea's sunscreen shelf was built on shared formulas, and it is about to find out what that transfer costs.
Comments