Retail giant Walmart launches four AI-powered agents

Walmart introduced new “super agents” meant to cut work for employees and customers alike. At its Retail Rewired innovation event, the company debuted four agents. Marty for sellers and suppliers, Sparky for shoppers, an Associate Agent for employees, and a Developer Agent.

Tariffs, inflation, and other cost pressures have raised doubts about household spending, pushing retailers to look for ways to keep sales moving. Some are betting on hands-on service led by store teams, while others are turning to artificial intelligence to streamline how people shop. Walmart falls into the latter group.

The four AI agents handle jobs such as payroll, paid time off, merchandising, and recommending items for specific occasions, bringing many tools together to simplify how people interact with the company.

“Having a plethora of different agents can very quickly become confusing,” Suresh Kumar, chief technology officer for Walmart Global, said at the event.

David Glick, senior vice president for Enterprise Business Solutions at Walmart, said the Associate Agent serves as “a single point of entry where any associate can find access to all of the agents we’ve built on the back end.”

“As you speak to it more, as you work with it more, it’ll know more about you”, he added.

Walmart is not alone in leaning into AI.

The shift comes as retailers search for ways to blunt rising costs for consumers and meet policy pressures.

During Amazon’s four-day Prime Day in July, generative AI usage climbed 3,300% year over year. Google Cloud AI also teamed with body-care brand Lush to visually identify unpackaged products, helping lower training costs for new staff.

Walmart is also investing in spatial and physical AI by building “digital twins” of its stores and clubs, virtual replicas used to monitor and manage operations.

With this approach, the company can “detect, diagnose and remediate issues up to two weeks in advance,” said Brandon Ballard, group director for real estate at Walmart US. The company says this work is paying off. “Last year, we cut all of our emergency alerts by 30% and we reduced our maintenance spend in refrigeration by 19% across Walmart US,” he added as quoted in a CNBC’s report.

Walmart uses AI to improve delivery time accuracy

“At its core, retail is a physical business,” said Alex de Vigan, CEO and founder of Nfinite, which produces large-scale visual data to train spatial and physical AI systems. “We’ve seen retailers use digital twins to reduce setup time for new promotions, reallocate labor more efficiently, and improve robotic picking accuracy, small gains that add up quickly when margins are under stress,” he said.

Although shoppers may not directly notice digital-twin work the way they would a tool like Sparky, the effects reach the customer experience.

“Better stock accuracy, faster site updates, and fewer order issues mean a smoother retail experience, even in a tighter economy,” said de Vigan.

Behind the scenes, Walmart is also applying machine learning to refine delivery-time predictions so customers have clearer expectations and operations run more efficiently.

On the consumer side, Sparky already builds carts based on an understanding of each shopper’s needs. Walmart is developing the agent so it can automatically reorder staples, aiming to ease the mental task of restocking.

For retailers, AI is one lever to offset a potential cooling in consumer demand. What remains to be seen is how a fully connected AI experience, online and in stores, will reshape how people shop over time.

Want your project in front of crypto’s top minds? Feature it in our next industry report, where data meets impact.