[COS IN Korea] “2026 Cosmetics Market Issues and Outlook Conference": AI as an Autonomous Expert and Co-Founder

This article is originally posted by 권태흥 기자(Kwon Tae-heung) on COS IN's December 1, 2025 Article. To read the full feature, please visit the link below:

‘2026 화장품시장 이슈와 전망 컨퍼런스’⑥ AI는 스스로 일하는 전문가·공동창업자

[코스인코리아닷컴 권태흥 기자]화장품 실무자라면 컨퍼런스에서 꼭 만나고 싶은 발표자가 바로 ‘AI가 바꾸는 K-뷰티의 성공 공식’을 설명한 트렌디어AI 천계성 대표다. 천 대표는 “최근 10년 사이 K-뷰티는 10배 성장했다. 눈여겨볼만한 점은 선진시장에서도 최대 수출국으로 올라섰다는 포인트다. 이는 한 지역의 트렌드가 아니라 전 세계에 새로운 뷰티 스

[Cosinkorea.com Reporter Kwon Tae-heung] For cosmetics practitioners, one presenter they most want to meet at conferences is Kei Chun, CEO of Trendier AI, who explained “the success formula of K-beauty changed by AI.”

CEO Chun began by saying, “Over the past 10 years, K-beauty has grown tenfold. A point worth paying attention to is that it has risen to become a top exporting country even in advanced markets. This means that it is not a trend limited to one region, but that it is creating a new beauty standard worldwide. Usually, when talking about K-beauty’s innovation, people mention speed or infrastructure. Recently, what is creating K-beauty’s unique competitive advantage are two elements: data and AI.”

Data is divided into △ the supply side, ‘detail pages’ that appeal through ingredients, efficacy, marketing keywords, and price, and △ the demand side, review data such as purchase reviews, pros, and cons. CEO Chun pointed out that until now, marketers have manually processed data one by one across dozens or even hundreds of competitor sites or multiple distribution channels. As a result, marketers have been chased by report analysis.

CEO Kei Chun explained, “Starting in May 2022 as the Trendier dashboard, we have now expanded to 30 distribution channels and produce more than 100 reports every month. What is interesting is that despite being a domestic B2B platform, more than one million reports are accessed and used for work every year. Marketers have begun to read trends in each market and propose various data-based formulas.”

With OpenAI recently announcing technology that enables direct connections with external apps, Trendier AI has also evolved. That is, in order to block AI hallucination errors and enable understanding of the beauty market’s unique context, an AI boot camp was opened last May.

CEO Chun said, “Initially, 100 teams participated and provided feedback, practical work was reduced to one-tenth, and automation elements were discovered. Last month, by opening Trendier AI linked with Trendier data and ChatGPT, analysis with context specific to the beauty market became possible.”

For example, when looking at the output from analyzing six months’ worth of data needed by Olive Young, examples such as brand names, actual reviews, growth rates, key ingredients, and the basis for buzz can be seen. This output is then questioned again as to whether it matches the marketer’s intent, and only when the answer is yes does it become connected to actual data that can be finally interpreted and analyzed.

CEO Chun proposed “a hybrid team of humans and AI when companies face difficult decision-making situations.”

On this day, CEO Chun showed the actual data utilization generation results of Trendier AI with the question, “Organize the popular efficacy + ingredient combinations promoted by domestic PDRN products and compare the ratio of positive and negative customer reactions.” He said, “Scientists can create all combinations of ingredients and efficacy, but predicting consumer reactions and sales volume after launch is a difficult task. If you think of it as a concept where this can be completed within a day even without receiving feedback after product launch, it makes sense.”

If asked about category selection when launching a new K-beauty product on Amazon, outputs such as ① discovering rapidly growing K-beauty keywords, ② detailed category segmentation and pricing, and ③ market validation through competition and demand analysis are generated. CEO Chun advised that because marketing makes it difficult to find the conditions for strong sales, the use of AI technology has become important.

He said, “Rather than being a decision made by a kind of tool, it is more accurate to accept AI as an HR (human resource) strategy.” AI is being used across major workflows—from R&D and product planning to market sales—and many companies are utilizing it. He emphasized that AI is now a virtual expert that understands its own context, sets goals, and works on its own, or even a co-founder, rather than merely handling repetitive tasks.

When companies face difficult decisions such as what to launch, how to sell, and where to expand, CEO Chun suggested that AI functions as a human–AI hybrid team.