At NRF, companies like Dick’s Sporting Goods and Sweetgreen talk up expanded uses of AI
At the NRF’s 2025 Big Show in New York City this week, AI continued to be a buzzword.
At last year’s event, companies like Google and Salesforce released new AI tools in time for the show. Meanwhile, big retailers like Walmart were starting to talk about new AI chatbots that had only just recently rolled out and were designed to address basic customer or employee inquiries.
This year, retailers were able to talk more tangibly about how they have incorporated AI and automation into their businesses over the last 12 months. There is still excitement for generative AI to improve the customer experience through chatbots and search recommendations, such as Google Agentspace. However, retailers seemed to be more focused on showcasing how they are using AI to solve challenges in business-critical areas like inventory planning, pricing and product R&D.
“Having attended in the last few years, ‘AI’ was thrown around as the be-all, end-all for all our problems,” one brand attendee who works for a big-box retailer told Modern Retail. “So it’s great to see finally some retailers are actually seeing tangible results from investing in these tools.”
Improving operational efficiency
Dick’s Sporting Goods was one retailer that used this year’s event to unveil a new AI tool: its AI-powered pricing model powered by Oracle Retail Labs. Dick’s Sporting Goods handles thousands of SKUs across its large-format stores and wanted to create a more predictive, accurate model to help forecast its markdown strategy going forward.
In the presentation, Juan Hernandez, director of omnichannel pricing at Dick’s Sporting Goods, outlined how the retailer is leveraging AI and machine learning to optimize its markdowns. That includes managing weekly discounts and sales, improving demand forecasting and insights into inventory shortages or delays.
“We make a lot of price changes, around 11 million SKU combinations every week,” Hernandez said. That also includes price tags on items whose prices don’t change, Hernandez said, which makes it more complicated to keep track of. “We are constantly adding new formats and stores, so we have to have a certain degree of scalability and flexibility.”
For instance, when Dick’s decided to promote a large assortment of golf apparel on the fly.
“We aggregated, recalculated all the demand and ran markdowns within 72 hours,” Hernandez said. “We did that in about a week or two.”
Elsewhere, AI-based robotics are starting to bear fruit. Sweetgreen CTO Wouleta Ayele discussed integrating automation and robotics into the fast-casual chain’s operations.
The company began testing its “Infinite Kitchen” tech in select stores, which has the ability to make up to 500 bowls per hour. Ayele said the concept is already freeing up team members to “move up front” of the store, allowing workers to interact more with customers. “So we’re very excited [for] the expansion of Infinite Kitchen and. It has been one of our innovative platforms.”
The limitations
Still, at major tentpole events like NRF, exhibitors are recognizing the saturation of AI tools on the market.
Joe Groves, senior vp of global sales at Centric Software, said the overwhelming promotion of “AI-powered” solutions at shows like NRF has become overwhelming for retailers trying to gauge what they need.
Groves said that for the past few years, many AI-focused solution vendors have popped up at NRF, “and they’re promising retailers the world with the use of AI.”
Centric Software sells product management software and was founded more than 20 years ago. As such, Groves said the company has certain advantages when building new tools — including client data from the past decade and that it employs hundreds of data scientists to help analyze it. “We are the system of record for how [our clients] plan, design, develop, source, manufacture and price all of their products,” he said.
One of Centric’s latest AI solutions helps fashion brands build a new style of an existing product — like releasing a different cut of jeans — by generating a list of all the parts and materials needed to make the new product. Clients include LVMH and Capri’s Michael Kors and Versace. “Our customers want to get to market faster, improve their margins faster, and optimize their inventory,” Groves said.
Still, retailers are acknowledging the limitations of AI in solving longstanding retail problems.
At Dick’s Sporting Goods, Hernandez said the new pricing model involves contextual data that teams can modify on the go based on real-world factors and events. “We’re not big fans of black-box models that take [data] and implement on their own — people are always going to want to have a say and control the decisions,” Hernandez said.