Artificial intelligence and its surrounding applications are just a tool. That’s a key point that many miss, and the reason why some fear that they’ll be replaced by machines. While some jobs will be lost, new ones will be created. And those who embrace AI could find the tech helps them do their jobs even better than before. Sourcing Journal spoke with management consultancy experts Robert Gorin and Michael Osment, both of Getzler Henrich & Associates, a Hilco Global company, about AI and retail. Gorin is a managing director and the consumer products practice leader, while Osment is a managing director and the analytics, AI and advance technology practice leader.

Sourcing Journal: There’s so much focus now on AI. How do you think it will impact the labor front, particularly in retail?

Robert Gorin: There will obviously be a lot of hiring in AI-related positions across retail, which will impact needed skill sets. The application of AI within retail will automate back-office routine processes, analyze purchasing and cost data to determine optimal new store locations, inform call center employees and use chatbots to allow for 24/7 customer service, as well as create and enhance in-store interactive technology applications. AI will also help retailers make use of available data to drive store staffing needs, enhance product design and set optimal inventory levels. New technology will further help anonymize data allowing retailers to make use of data that is difficult to use today because of privacy concerns and other reasons. Access to these new forms of analytics will help retailers maximize the lifetime value of the customer. Given the expected across-the-board impact of AI, retail staff will need new IT-based skills in combination with their marketing, product and customer service talents. Understanding AI alone won’t be enough to drive success. Knowledge of the customer experience will be critical to help companies stay client-focused and enhance the customer experience across the spectrum of customer interactions.

SJ: How would AI help on the customer experience front?

RG: New technologies being deployed to enhance customer experience, such as fitting room smart mirrors or other interactive displays, will be able to better tailor the shopper experience by combining in-store and online preferences into a consistent shopper identity that follows the shopper across their omnichannel journey. Interactions with technology, bots, and staff appear effortless and seem as if the store itself is responding to and interacting with the shopper and their mission on that day. It will also allow for consistent, knowledgeable, and personalized customer experiences, regardless of the sales channel being employed. This application of AI creates a level of personalization not previously experienced, whether it be in product design, email marketing or other customer communications. It will make the customer feel understood, which creates a strong brand-to-customer bond.

SJ: But haven’t retailers been using data for years? What’s different now?

Michael Osment: The big difference is the rise of generative AI and large language models, especially private ones that have been trained on the company’s proprietary data. A large language model, such as GPT, is a type of artificial intelligence model that is trained on large amounts of text data to understand and generate human-like language. These models utilize deep learning techniques to process and generate responses based on input prompts or contexts. These systems are essentially predictive engines that are continually evaluating and weighing probabilities around what the shopper will do next or ask next. Big data and its applications such as using loyalty data for recommendation has been around. It’s just going to get a lot smarter, faster and better at predicting what comes next. Also, given that AI makes the data more accessible and easier to parse, companies both big and small can reap the benefits. Because of this, more and more companies are focusing on data analytics, and it is thus becoming more mission critical for competing companies to truly leverage their data.

SJ: So how can AI help sales staff, as well as the merchandising teams, with the curation of the assortment mix on the sales floor?

RG: Building a private large language model and training it on all the data available including location data, video analytic data, assortments, plan-a-grams and the rest of the rich data available to a retailer—either their own or purchased data—really will allow them to build a 360 degree model of their shoppers generically, but also specifically to the people in the store or on their way. This can support real-time AI determined dynamic pricing, inventory movement within stores, and even availability of colors on a shelf. Using all of the omnichannel data available on a given customer will help have the most accurate and desired products on the sales floor. It will allow the sales representative to be responsive, knowledgeable and personalized in their customer interactions. It will speed up and smooth the shopping experience as it creates convenience for the customer. AI should also help optimize inventory which will limit stockouts—and overstocks—and reduce potential shopper disappointment when they can’t purchase a desired item.

SJ: What is the skill set that retailers are looking for?

MO: There really isn’t much changing about the key characteristics of the optimal retail associates that the retailers want. They want friendly, personable and attentive people who care about the customer experience and enjoy working with customers. The AI will assist, but that assistance should be unobtrusive, unobserved by the shopper. As for people in product design, inventory management, marketing and purchasing, the retailers need staff who can review the analysis, understand the narrative behind the data and can apply it when making decisions. Key questions when parsing through the data include what has the company learned about its customer and what does it need to do to adjust or enhance its business. Being able to translate the story that the data is telling you is critical.

SJ: Where does this person get the experience? Is it from coursework? Or does the person have to be a merchandiser or marketer first?

MO: It will always be the case that having a solid understanding of how the business operates is the best way to start. Comprehending the business processes, metrics and goals makes for a better employee generally. Having a curiosity about technology and an interest in turning a job in retail into a career as a merchant, buyer or retail operator is critical. The employees who have a passion for retail will embrace AI because they will see firsthand how it can make their stores more attractive and deliver an amazing customer experience. A key pitfall to avoid is focusing on the technology first and not on the basics of retail or customer experience.

RG: As an example, we have a client with a young salesperson who understood how buyers at the retail level shop, and she became interested in AI. At night, she went to school and got a degree in it. Because she was the top salesperson, it was a hard decision to take her out of the field and put her in charge of the company’s AI efforts. But now her impact is much broader than it was before. Now, she is informing and educating the entire salesforce.

SJ: What do you think will be the biggest use of AI in the supply chain?

RG: Initially, it’ll be the easy things where there is a lot of data. These will be areas like replenishment, assortments and allocations of product. Today, these calculations are based on data from previous weeks, quarters or years and so are always lagging. Very soon, retailers will begin rolling out private GPT’s that incorporate this data, but add in social media feeds, trends in competitive industries, supplier specific data, and more to predict impacts not seen in the traditional data. Big Data, Machine Learning and Gen AI will all combine to enhance supply chains and ensure that inventory at the size, style and color level exists when and where it is needed. Further, linking this information to other AI models that offer customers recommendations on what they may prefer to buy next will drive consumers toward what is on-hand based on all available data. This again should increase revenue, limit overstocks, and create happier customers. AI will also help companies diagnostically with the basics such as driving costs out of each step in the supply chain.

SJ: Do you foresee the creation of new jobs for these analytical tasks or existing staff upskilling their current skill set via training?

MO: The impact on jobs of Gen AI will be primarily on the sorts of white collar “knowledge workers” that have previously been somewhat insulated from job cuts. The question is whether you can take the people in your legal, accounting, finance, customer service, IT and other areas and re-deploy them as AI developers. There will be a new class of employees needed in areas like “AI Model Trainer” and “Prompt Engineer,” but it is unlikely that the volume needed of these folks will be anything close to the attrition among the knowledge workers. Retailers will continue valuing employees who embrace continuous learning and upskilling. And those employees will find lots of opportunities as AI becomes ever more common across the retail organization.

SJ: What do you think is the greatest hope for AI in fashion retail and where do people and retail jobs fit in with the new technology?

RG: AI and advanced technology will transform retail supply chains. Today, when you order a book, it is printed and shipped within a few hours. Except for the most popular titles, there is no warehouse full of books waiting to ship. Expect similar capabilities to be applied to the world of fashion. I spoke with a company the other day that was finalizing a new process to take orders and manufacture apparel to order based on actual demand. It optimizes production by checking things like which seamstress has black thread already loaded, who has the capability and experience to sew shoulder seams and where are the bolts of black cotton that are already in place. This will eventually go backward in the process all the way to dyeing the fabric a specific color to match an exact garment that someone just designed, much like you mix paints. Your only raw inputs will be undyed fabric and thread, but you’ll still be able to produce any garment in less than 12 hours. AI also will enable people to more quickly recognize and adjust to the start of new trends: from where in the country they are buying, to what styles and colors they are buying, and to how much consumers are willing to spend for it. Add that to AI’s ability to forecast what customers are likely to purchase in the future—that’s the real golden promise of AI. As for people and retail jobs, humans aren’t going to go away. AI is still a machine and at the end of the day, people will drive the creative spark and understanding that is layered on top of AI. The power comes from AI plus the human perspective. It’s not one or the other.