kbeautytmi › AI citations

When people ask AI about K-beauty,
whose name comes back?

Every week we ask 4 AI engines (Perplexity · Gemini · OpenAI · Claude) the same 75 K-beauty questions — best toner for sensitive skin, best Korean sunscreen, and so on — and count which brands the answers actually cite. Citations ≠ sales. This is the discovery layer forming before anyone reaches a shelf. Measured 2026-10-12

The leaderboard

Violet bar = how often AI names the brand. The chip = how many of our 20 retail channels it actually charts on — the gap between fame and shelf is the story.

1KLAIRS 74on 3 channels
2COSRX 47on 9 channels
3Anua 37on 14 channels
4LANEIGE 32on 5 channels
5Dr.Jart+ 30on 2 channels
6medicube 28on 13 channels
7Paulaschoice 28on 2 channels
8ETUDE 24on 3 channels
9AESTURA 22on 7 channels
10Torriden 19on 4 channels
11Eucerin 17on 1 channel
12SKIN1004 17on 8 channels
13MEDIHEAL 16on 6 channels
14The Ordinary 16on 4 channels
15ILLIYOON 15on 4 channels
16INNISFREE 15on 3 channels
17BIODANCE 14on 9 channels
18Avene 13not on our shelves
19NATURIUM 12on 3 channels
20Glow Recipe 11on 2 channels
21Haruharu Wonder 10on 7 channels
22Pyunkang Yul 10on 2 channels
23ROUND LAB 10on 10 channels
24Abib 9on 5 channels
25Beauty of Joseon 9on 10 channels
26Bioderma 8on 1 channel
27Dr. Althea 8on 12 channels
28Neogen 8on 1 channel
29Purito 8on 5 channels
30BENTON 7not on our shelves
31Isntree 7on 3 channels
32numbuzin 7on 9 channels
33Sulwhasoo 7on 3 channels
34SKINFOOD 6on 3 channels
35SOME BY MI 6on 4 channels
36celimax 5on 8 channels
37goodal 5on 2 channels
38Real Barrier 5not on our shelves
39VT Cosmetics 5on 3 channels
40belif 4not on our shelves
41BRINGGREEN 4on 2 channels
42I'M FROM 4on 3 channels
43It's skin 4not on our shelves
44JUMISO 4on 1 channel
45MISSHA 4on 2 channels
46mixsoon 4on 4 channels
47REJURAN 4on 2 channels
48BANILA CO 3on 3 channels
49HANYUL 3on 1 channel
50PEACH & LILY 3on 1 channel
51THE FACE SHOP 3on 3 channels
52TONYMOLY 3on 3 channels
53TOSOWOONG 3not on our shelves
54AROMATICA 2on 1 channel
55beplain 2on 5 channels
56Dieux 2on 2 channels
57Holika Holika 2on 2 channels
58Hyggee 2on 1 channel
59IOPE 2on 2 channels
60JUNGSAEMMOOL 2on 3 channels
61Peach Slices 2not on our shelves
62S.NATURE 2on 5 channels
63TIRTIR 2on 1 channel
64TOCOBO 2not on our shelves
65APLB 1on 2 channels
66AXIS-Y 1on 4 channels
67Centellian24 1on 3 channels
68d'Alba 1on 5 channels
69Demaf 1not on our shelves
70Dermatory 1not on our shelves
71Dr. Melaxin 1on 1 channel
72easydew 1not on our shelves
73EQQUALBERRY 1on 4 channels
74FRUDIA 1not on our shelves
75Hada Labo 1not on our shelves
76Heimish 1on 2 channels
77Hero Cosmetics 1on 3 channels
78KSECRET 1on 2 channels
79LINDSAY 1not on our shelves
80Mamonde 1not on our shelves
81MEDIPEEL 1not on our shelves
82Meditherapy 1on 1 channel
83NIDA 1on 1 channel
84Petitfee 1on 1 channel
85Rohto Mentholatum 1on 1 channel
86ROVECTIN 1on 1 channel
87SATURDAY SKIN 1on 1 channel
88SUNGBOON EDITOR 1on 1 channel
89Whipped 1not on our shelves

Week over week

2026-07-06
№1 COSRX 40
2026-07-13
№1 Anua 36
2026-07-27
№1 COSRX 49
2026-08-03
№1 COSRX 56
2026-08-10
№1 COSRX 46
2026-08-17
№1 COSRX 59
2026-08-24
№1 Anua 13
2026-08-31
№1 COSRX 56
2026-09-07
№1 COSRX 59
2026-09-14
№1 COSRX 58
2026-09-21
№1 COSRX 50
2026-09-28
№1 COSRX 56
2026-10-05
№1 COSRX 49
2026-10-12
№1 KLAIRS 74

The trend line is the product — the archive deepens every Monday and Thursday.

What Copilot cites on generic queries

From Bing's AI Performance data — non-brand queries only.

"what do toner pads do" 30 citations · 26.79% share
"how to have glass skin" 22 citations · 20.56% share
"difference between dark circles and under-eye shadows" 18 citations · 16.67% share
"caffeine eye masks refrigerated effectiveness" 16 citations · 17.39% share
"dermatologist-tested meaning shower gel" 15 citations · 39.47% share
"glass skin skincare routine steps" 14 citations · 23.33% share
"milky toner skincare definition" 14 citations · 19.18% share
"glass skin routine steps" 13 citations · 22.41% share
"how to tell if moisturizer feels sticky or heavy before buying" 12 citations · 42.86% share
"dermatologist-tested meaning skincare" 12 citations · 36.36% share
"fragrance-free vs naturally scented skincare sensitive skin" 9 citations · 11.25% share
"cica exosome" 8 citations · 18.6% share
"dermatologist-tested skincare meaning" 6 citations · 28.57% share
"caffeine eye masks refrigerated benefits" 5 citations · 21.74% share
"should you refrigerate eye masks for puffiness" 5 citations · 45.45% share
"ph-balanced cleanser meaning importance" 4 citations · 17.39% share
"hypoallergenic deodorant guarantee meaning" 3 citations · 11.54% share

Why this matters

Shoppers increasingly ask AI before they search, and AI answers concentrate on a handful of names. A brand that dominates retail shelves but never gets cited is invisible in the next discovery channel — and a brand AI loves that hasn't hit mass retail is a signal buyers can act on. Cross-reading this leaderboard against our retail ladder is the whole point: in 2026-10, the top of the AI leaderboard and the top of the retail-reach ladder were the same brands — evidence the discovery layer and the shelf are moving together.

Method: identical prompts to 4 engines, brand mentions extracted from answers with token-boundary matching against our chart-brand dictionary. Reprocessed brand-neutrally by kbeautytmi. AI citation ≠ endorsement, quality, or sales.
What this measures, and what it does not. These 65 questions were not written to survey K-beauty evenly. They lean hard toward sensitive and barrier-focused skincare — 27 of the 65 ask about sensitive skin, and three name a single ingredient. A brand that sells into that space will rank higher here than it would on a question set built for the whole category. Read this as who AI cites for these questions, not as who AI cites most in K-beauty overall. The raw engine measurements are shared from an external skincare GEO engine and reprocessed here; the brand dictionary and scoring are ours. We are building our own question basket to replace this one — until then, treat the ranking as directional. How we measure.