After controversies over inappropriate marketing by some corporations, a red alert for marketing risk management has been issued across retail and fashion. While buzz and trends matter, blocking controversies in advance—such as political, social, and gender issues—has emerged as a core task for marketing. Some fashion platforms are creating dictionaries that organize online community memes and neologisms, and are expanding the use of artificial intelligence (AI) models to reduce advertising model risk.

An AI model wears a Louis Vuitton Alma BB bag. /Courtesy of Louis Vuitton website

On the 5th, according to related industries, fashion platform ably has produced a "community-er dictionary" and is sharing it with all employees. It turns memes, neologisms, slang, gender-related expressions, and words with political and social context used in online communities into databases (DB) so they can be reviewed in advance when creating marketing copy or social media (SNS) content.

Fashion platform Musinsa has also advanced its marketing review system. Rather than merely having the department in charge organize and share materials, it has built and is operating a system that can check marketing copy or content in advance. A Musinsa official said, "We have created a process that allows a prior check for associations with specific issues, memes, or similar expressions when marketers or planners enter content into the system," adding, "We have continually advanced the related review framework since past issues and are once again inspecting it recently."

Because fashion platforms are sensitive to online trends, there is a relatively higher chance that certain expressions or buzzwords will unintentionally escalate into controversy. In particular, platforms targeting a specific gender are even more sensitive to gender issues and are strengthening related review procedures.

Behind the industry's recent jitters is the controversy over Starbucks Korea's "5·18 Tank Day." Anxiety has grown that certain dates, numbers, or phrases can be interpreted with social or political meaning regardless of corporations' intent, potentially leading to boycotts or damage to brand image. Within the industry, some are also raising the need to build a so-called "risk calendar" that organizes socially sensitive dates such as the Sewol ferry disaster, the May 18 Democratization Movement anniversary, and the deadly crowd crush in the nightlife district of Itaewon.

A fashion industry official said, "Even now we sometimes share monthly key dates and related issues, but we feel the need to build this into a system."

An AI campaign photo for Jill Stuart New York Women by LF Corp. /Courtesy of LF Corp.

Marketing risk management is also expanding into the realm of advertising models. In the past, stars or influencers were often used as models, but there were not a few cases in which brand image took a hit if controversies arose over private life or remarks. Analysts note that growing adoption of AI models in the fashion industry recently also reflects this demand for risk management.

VOV, a women's wear brand operated by Shinsegae Tom Boy, unveiled the AI virtual model "Victoria," and LF Corp.'s Jill Stuart New York Women is running an AI-based campaign. Indian, a men's wear brand of the Saejung Group, is also using a senior AI model in lookbooks and marketing. Global luxury brand Louis Vuitton is likewise producing some product-wear images with Generative AI.

Until now, the advantages of AI models have been cited as expense savings and production efficiency. They reduce the process of recruiting real models, shooting, and retouching, and allow quick responses as seasons change. Recently, the potential for near-zero-risk management has also emerged as a highlighted benefit.

A fashion industry official said, "In the past, we considered AI models for model fee savings and production efficiency, but recently there's a perspective of viewing them in terms of brand risk management," adding, "Real models can face controversies over private life or unexpected variables, but AI models have very little such risk, which is an advantage."

※ This article has been translated by AI. Share your feedback here.