Dialogue with Sylvia Fu, Global AI Transformation Leader at Edgewell: Activating the World with "Chinese Speed" and Remodeling the Future of a Consumer Goods Giant with AI
In the current era when the global consumer goods industry is undergoing profound changes, Sylvia Fu, with her unique cross - cultural perspective between the East and the West and the execution ability of "Chinese speed", is writing a brand - new story about how AI is reshaping traditional industries. From 2022 to 2025, as the Vice President of Edgewell Personal Care in Greater China, she delivered an impressive report card. For three consecutive years, the sales and profit growth in Greater China ranked first globally, and she successfully incubated one of the most commercially explosive new products in the group's history in the Chinese mainland. At the end of last year, Sylvia officially took up the position of Global AI Transformation Leader at Edgewell, reporting directly to the global CEO, which marks that Edgewell has upgraded AI from a "single - point tool" to a "core strategy".
At the end of March 2026, she just returned to New York after attending the retail technology summit Shoptalk Spring in Las Vegas. 36Kr finally had the opportunity to talk with her. Standing at a new starting line, how will she lead this bottom - up cultural change? How is the US retail industry and consumers being reshaped by AI? And what does this mean for the globalization of Chinese enterprises? The following is the transcript of this exclusive interview.
Sylvia Fu, Global AI Transformation Leader at Edgewell
01. Strategic Orientation - Reject "AI for AI's Sake", and Strengthen the "Business - Oriented" Approach with Three Pillars
36Kr: Sylvia, as the Global AI Transformation Leader, could you tell us what you plan to do next? Where is Edgewell currently in its AI practice?
Sylvia Fu: In the strategic report just presented to the global board of directors, we elaborated on a very clear transformation framework. In the upcoming AI layout, our core concept is extremely clear - "Operator - First".
Many companies' AI transformation is driven by pure technology departments, which are prone to fall into the trap of "technology for technology's sake". However, I used to be in charge of business and was responsible for the P&L (Profit and Loss Statement). Therefore, the only criterion for me to evaluate AI is whether it can be translated into real commercial leverage: that is, to drive top - line growth (increase sales, sell more), improve organizational agility (be faster), optimize bottom - line efficiency (increase profits, reduce costs), while taking into account responsible data governance and cultural construction.
Based on this concept, we have built a framework of "Three Modules". Let me briefly share it:
The first module: Integrate AI into daily workflows. AI transformation is essentially a transformation of "people". The goal is to enable employees in every position globally to use large models such as Copilot and ChatGPT as regular "digital assistants". At the same time, we are developing exclusive "AI Agents" for functions such as finance, procurement, and HR to reshape workflows. We have selected more than 100 cross - departmental backbones globally to form the "AI Advocates Club" as a cultural communication engine, and through peer - to - peer learning, we aim to start a prairie fire of organizational transformation. To address the concerns of the business side about data security, we will release the authoritative "Safe Harbor Guidelines" in the middle of the year to define compliance boundaries and encourage everyone to work smartly within the safe zone.
The second module: Strengthen the data and technology foundation. Facing the common data silos in multinational enterprises, we have adopted a very practical "Dual - track" approach: on the one hand, we will not wait for the so - called perfect data. We will give priority to using the currently available internal and external data to quickly launch pilots and verify the commercial ROI; on the other hand, we will continue to invest in the infrastructure of the underlying system data lake, unify the definition and granularity of global data, and pave the way for the large - scale explosion of AI.
The third module: Focus on core business scenarios. What I fear most is the "follow - the - trend" model - creating a chatbot one day and an image generator the next, which seems lively but has little real help for the business. Our approach is to first identify the major business pain points of the group, and then assign a business - responsible team and a technology - responsible team to each scenario. The two teams work together. In this way, we can ensure that AI is really solving problems rather than creating new complexities.
These three pillars actually form a flywheel - the data infrastructure provides ammunition for scenarios, scenario applications create profits for the business, and the AI culture with full - staff participation makes this flywheel spin and spin fast.
After talking about the strategy, let me talk about the actual progress. Frankly speaking, the speed is even faster than I expected, which really surprises me -
At the R & D innovation end: We have cooperated with the global AI open - innovation platform Halo to significantly compress the cycle of finding breakthrough technologies such as PFAS - free coatings and new UV - absorbing compounds from several months to just a few weeks, with an efficiency increase of 90%. In the blade replacement coating project, in the past three months, we received 24 technology proposals through the AI network and quickly identified 9 high - potential partners, and now we have entered the in - depth technical due - diligence stage.
At the marketing insight end, just in February this year, we did something really cool with the North American team - we used a large language model to analyze more than 190,000 social and review data points and defined 5 new consumer segments for the sunscreen category. These insights are directly guiding the product development and marketing directions of our two brands, Banana Boat and Hawaii Tropic. In the past, conducting such in - depth consumer insights would take at least three months and cost millions. Now? It's done in a few weeks, and the granularity is finer.
At the global operations end: The European supply chain team is using machine - learning models to advance 5 active workflows, including Center of Gravity (CoG) Simulation and demand - sensing prediction. The newly developed AI tools at the global legal end have covered 6 major scenarios such as contract review, saving 25% of working time.
02. Insights on the Frontier - The "Trust Paradox" between the US Retail Industry's Full - scale Embrace of AI and Consumers
36Kr: You've spent most of the past six months in the US. From your first - hand perspective, what qualitative changes have occurred in the attitudes of the US retail industry and consumers towards AI?
Sylvia Fu: The most powerful impact I've felt in the past six months is that the attitudes of US retail giants towards AI have completely crossed the early "Proof of Concept (PoC)" stage and fully entered the in - depth reconstruction of the core value chain.
We can clearly see that the giants are building an "AI - first" underlying ecosystem. David Guggina, the newly appointed CEO of Walmart US, clearly pointed out that "Agentic AI" is being widely embedded in Walmart's daily operations and checkout processes. Best Buy is using generative AI to provide real - time emotional analysis for tens of thousands of customer service representatives, freeing up employees' energy to build "customer resonance"; Costco is deeply applying AI to predictive demand insights and seamless checkout.
The latest forecast from EMARKETER is extremely shocking: The US e - commerce sales driven by AI platforms will soar from $5.4 billion in 2025 to $144.5 billion in 2029 (accounting for about 9%). In an aggressive scenario, this figure could even reach as high as $225 billion. This means that every brand must re - think its new business opportunities in the new AI ecosystem.
Brands under Edgewell
However, on the other side of the coin, there is a very inspiring phenomenon on the consumer side - what I call the "trust paradox". According to a survey released by Publicis Commerce in January 2026, up to 64% of US consumers have used AI shopping tools, and their shopping behavior is shifting from simple "keyword search" to a "problem - solving" mode based on complex intentions. However, the paradox is that although 92% of people think AI is helpful, only 52% of them really trust AI product recommendations.
Interestingly, consumers are most likely to trust AI recommendations only when they are already familiar with a certain brand; moreover, 60% of people's first reaction after receiving an AI recommendation is to "cross - verify in real - human communities such as Reddit".
This has taught us a very profound strategic lesson: In the AI era, "High Tech" must be paired with "High Touch". When AI makes product comparison effortless, the trust, emotional connection, and real human experience accumulated by the brand itself are the ultimate moats to break the trust barrier and drive conversions.
03. Core Battles - Four Key Directions for Edgewell's AI Development
36Kr: Facing this rapid evolution, what will be Edgewell's "core battles" in AI next?
Sylvia Fu: We are currently closely evaluating with some very powerful partners. There are mainly four directions, and I'll talk about them one by one:
First, intelligent supply chain and demand forecasting. This is the lifeline of fast - moving consumer goods. To be honest, traditional SKU - level forecasting can no longer keep up, especially when new products are launched, and historical data has little reference value. We are researching a "data weaving layer" approach. Instead of changing the existing systems or making major overhauls, we will add an AI intelligent layer on top to break through data silos and then use digital twins for more accurate forecasting. Even if the forecasting accuracy only increases by 10 percentage points, it will be a huge improvement for our sales and inventory health.
Second, data infrastructure and system readiness. No matter how smart the AI model is, if the data you feed it is messy, the results will also be messy. Our most practical approach now is to first make the SAP system truly "AI - ready". The cleaning and construction of the data lake are the mid - term strategic priorities. This task is our mid - term focus, and if we don't do it, all future AI applications will be castles in the air.
Third, brand discovery in the AI era - from SEO to GEO. I really want to talk more about this because it is really changing the game rules. Now, 60% of Google searches result in zero clicks - consumers no longer "browse links" but directly want "answers". When someone asks ChatGPT "What sunscreen is good for sensitive skin?", if your brand is not in the AI's answer, you will disappear from the digital shelf. So we are constantly testing GEO (Generative Engine Optimization), specifically converting product information and consumer reviews into a structured format that large models can "understand", so that AI can actively recommend our brand when answering consumers' questions. This battle has just begun, and those who act first will have an advantage.
Fourth, AI - accelerated brand content and product innovation. In China, AI has helped us produce more than 2,000 videos on e - commerce platforms every month. Next, we will expand this ability to North America and Europe - for the large - scale production of product pictures, videos, and social content, while ensuring local - level quality. At the product innovation end, we are also using AI to connect the entire process from consumer insights to concept verification to product launch. The analysis of 190,000 data points of North American consumers mentioned earlier is an example, and the AI - based technology search by Halo at the R & D end is another example. In the future, these two aspects need to be connected to truly compress the innovation cycle.
04. "Agentic Commerce" is Here - How AI is Reshaping Consumers' Shopping Methods
36Kr: You've mentioned "Agentic Commerce" several times. Could you elaborate on how it will reshape the future retail form?
Sylvia Fu: "Agentic Commerce" has truly reached an inflection point this year. Let me use a timeline to illustrate this acceleration: In just half a year from September 2025 to March 2026, ChatGPT launched the Instant Checkout function, Amazon launched the Rufus shopping assistant, Google released the Universal Commercial Protocol (UCP), and Microsoft and Perplexity also entered the game one after another. All technology giants are frantically competing for the "next - generation AI shopping entrance".
In the future, there will be three types of consumers in retail: traditional human consumers, hybrid consumers who use AI to assist in shopping, and fully autonomous AI agents. The most interesting group is the middle one - they have one foot in the human world, valuing experience, personalization, and values; and the other foot in the AI world, using conversational search and multi - modal interaction. Currently, the most active part of AI shopping is the "help me choose" stage, but it is expanding to the entire shopping journey.
For brands, a very important signal is that consumers have changed from "searching for keywords" to "asking questions". They no longer enter "sunscreen SPF50" but ask AI "I have sensitive skin that gets sunburned easily. What sunscreen is good for going to the beach?" This is a completely different logic. Traditional SEO cannot solve this problem. You must do GEO - make AI understand your products and recommend your brand when answering questions. If AI doesn't recommend you, your brand may "not exist" in the eyes of consumers. This is why we list "winning brand discovery in the AI era" as one of Edgewell's four core opportunities.
05. Insights for Going Global - How Chinese Brands Can "Win Globally with Intelligence" in the AI Era
36Kr: As a person who operates across the East and the West, how do you view the significance of AI for Chinese enterprises going global?
Sylvia Fu: Chinese enterprises going global have now entered a new stage - it's no longer enough to just expand in scale, but to combine "value construction" and "technology - driven" development. In this process, AI is not just a bonus; it is actually the "core operating system" that supports global operations.
What used to be the biggest headache when going global? It was difficult to balance scale and localization. If you wanted to expand, you didn't have the energy to go deep; if you wanted to go deep, you couldn't expand widely. Now, AI has changed this equation - with capabilities such as multi - language generation, predictive analysis, and consumer insights, you can provide experiences that truly fit the local culture in each market at a very low marginal cost.
There is also a point that many people ignore: compliance. The EU AI Act and various countries' privacy - protection regulations may seem like constraints, but if you can systematically embed "trust design" into your products and operations, compliance will actually become your competitive advantage. Consumers will think "this brand is reliable", which is very valuable in overseas markets.
Chinese enterprises really have natural advantages in application innovation and agile iteration. As long as they inject "high empathy" into "high technology", they are fully capable of leading in this wave of globalization.
36Kr: Based on your practical experience in the Chinese market and your current global perspective, if Chinese brands want to go global to the US and truly build brands and channels, what is the most crucial thing? And how can AI help?
Sylvia Fu: I really have a lot of feelings about this question because I've come all the way from the Chinese market. To be honest, the strategies in the US are very different from those in China. I've seen many Chinese brands still using the domestic approach, and some detours can actually be avoided. I think there are three core things: how to establish a brand, how to open up channels, and how to establish a