Walmart Is Giving Its Associates Super Powers. Suppliers May Have More to Gain Than They Think.
At MIT Technology Review’s EmTech AI summit on April 22, EVP and Chief People Officer Donna Morris said Walmart intends to bring some level of AI skills to each of its 2.1 million associates over the next several years. According to CIO Dive’s reporting on her remarks, role-specific certifications are already open to the 1.7 million workers in the U.S. and Canada through the company’s internal training platform, Squiggly, with the program built in partnership with OpenAI and Google. Morris named two concrete use cases for front-of-house workers: faster stockroom item location, and real-time language translation with customers who speak a different language.
Most of the supplier conversation about Walmart’s AI rollout over the past 18 months has lived on digital surfaces. Sparky, Marty, catalog data, retail media. The associate transformation has been treated as an internal HR story. That framing misses something, and 1P suppliers in particular should look again. The store associate is the person who actually touches your product, your shelf, and your shopper. An associate equipped with agentic AI tools is a different partner than the one your field organization, whether in-house, broker-led, or run through a national retail services agency, has been calling on for the past decade. (This is largely a 1P story. Marketplace sellers without product on Walmart shelves can read on for context, but the operational implications below apply to suppliers with brick-and-mortar distribution.)
The Phantom Out Problem Just Got Smaller
The most underrated detail in Morris’s EmTech remarks is the stockroom search use case. Front-of-house associates locating items in the back faster sounds like an internal efficiency story. For suppliers, it’s an on-shelf availability story.
Every category leader knows the gap between system-on-hand and on-shelf availability. The item is in the building, the system says it’s in stock, but the shelf is empty because no one pulled it forward. Phantom outs cost suppliers real volume, particularly in high-velocity categories where a one-day stockout doesn’t trigger a replenishment alert but does trigger a lost trip. Suppliers have spent years working this problem through better case-pack design, modular optimization, third-party shelf audits, image-recognition tools that scan store conditions, and pleading with replenishment teams to look at lost-sales data more aggressively. None of those levers do anything if the associate can’t find the case in the back.
Walmart has been chipping at this problem in apparel for four years. Per the company’s June 24, 2025 announcement on associate AI tools, VizPick, Walmart’s AR-based inventory tool, has been helping associates find merchandise to stock on the sales floor since 2021, and the newly enhanced version pairs RFID with AR to visually guide associates to the specific items that need to move from backroom to shelf. The category was chosen for a reason. Walmart’s announcement identifies apparel as one of the most challenging categories to manage in retail because of high volume, fast turns, and constant movement. Suppliers should expect the same friction-reduction approach to extend to other categories as the broader agentic AI rollout reaches more associates, building on a base of more than 900,000 weekly users already running over 3 million queries per day through the existing Ask Sam voice assistant per the same Walmart announcement.
For suppliers, the practical floor on OSA rises across stores as that friction comes down. Suppliers who have been losing share points to phantom outs in specific categories should expect that erosion to ease. Walmart isn’t solving the problem on suppliers’ behalf. The friction in the associate’s workflow is being removed, and OSA improves as a byproduct. The supplier-side action is to actually measure it, and to revisit the ROI on the third-party shelf intelligence and store audit investments that were sized to a different baseline. Categories where lost-sales attribution has been muddy because the inventory data didn’t match on-shelf availability will start producing cleaner signal. That signal is worth pulling into the next JBP conversation as evidence of category responsiveness, not as a complaint.
What In-Store Coverage Is Actually For Is Changing
Most CPG in-store organizations spend a meaningful slice of their store visit time educating associates on basic product facts. New flavor, new pack size, what the claim on the front of the package actually means, how this item differs from the one next to it on the shelf. That work matters because an associate who can answer a shopper’s question converts a consideration into a purchase. For most suppliers, that work is delivered through a mix of in-house field teams, brokers, and retail services agencies, and the scope of work negotiated with those partners reflects assumptions about what associates know coming in.
If associates can pull product information through an AI agent in seconds, the marginal value of a rep delivering basic product facts drops. Ask Sam already handles more than 3 million associate queries per day, per Walmart’s June 24, 2025 announcement, and the EmTech rollout extends agentic AI capability across the broader workforce. The agent can deliver basic facts. What the agent can’t do is build a relationship with the department manager, walk the section with the associate to spot a planogram drift, secure a secondary placement for a seasonal push, or notice that the endcap two aisles over is a better fit for the new launch than the one in the planning packet.
Suppliers should reexamine whether their in-store scope of work is still aligned with what associates actually need from a visit. The teams that get this right will redeploy time from product education toward higher-leverage work. Relationship building with department managers. In-store merchandising audits. Securing incremental space. Gathering shopper insight that the agent can’t capture because it never leaves the system. That redeployment is a conversation suppliers need to have internally and with their retail services partners, because the call cycles, visit minutes, and reporting deliverables in most existing contracts were built for a pre-agent associate base. The suppliers who don’t have that conversation will spend the next two years paying for work the agent already did before the rep walked in the door.
Execution Variance Across Stores Should Narrow
There has always been a gap between flagship Walmart stores with veteran associates and lower-volume stores with newer hires. Veteran associates know categories. They know which items move on weekends, which displays get knocked over, which shoppers come in asking about which brands. New associates in newer stores don’t have that institutional knowledge yet. For suppliers, that variance has shaped everything from where to test a new item, to how aggressively to fund a regional rollout, to which stores get the demo budget, to how retail services coverage is weighted across the store list.
AI-supported guidance compresses that gap. An associate with limited tenure operating with on-demand access to product information, store maps, and process guidance is working from a different baseline than the same associate would have had two years ago. The implication for suppliers is that execution variance across stores should narrow over time. New item launches that historically showed wide velocity dispersion between top-quartile and bottom-quartile stores should see that dispersion compress as associate capability evens out. Test markets become more representative of national rollout performance. Trade dollars and coverage hours that used to be allocated heavily to flagship stores can be redistributed with less risk.
This is a forward-looking read, not a measured outcome. Suppliers should watch for it in their own data over the next several quarters and adjust accordingly. The teams that recognize the variance compression first will allocate trade dollars and coverage investment more efficiently than competitors who keep funding the old playbook.
Translation Is a Velocity Lever Hiding in Plain Sight
The translation use case Morris named at EmTech is more strategically significant than it sounds, and Walmart has been more specific about the capability than the headline coverage suggests. Per the company’s June 24, 2025 announcement, the real-time translation feature is available in 44 languages, supports both text-to-text and speech-to-speech formats, and is enhanced with Walmart-specific knowledge so that house brands like Great Value translate properly rather than being mangled by a generic translator. That last detail matters. A capability that handles brand names and Walmart-specific retail vocabulary is one that suppliers can rely on for accurate product representation, not just rough conversational handling.
In stores with significant non-English-speaking shopper bases, a supplier’s new item velocity has historically depended partly on whether the associate could explain the product to a shopper who didn’t share a language with them. Categories where associate recommendation tends to carry weight, including areas like HBA, baby, and the international foods aisle, are the ones where suppliers should consider the implications most closely. The conversion loss from language friction has rarely been attributed to its actual cause in supplier velocity reporting. Real-time translation between an associate and a shopper changes that math. Suppliers with products that benefit from associate recommendation, particularly in immigrant-dense trade areas, should see the recommendation friction ease. The supplier-side action is to look at velocity data by store demographic and ask whether items that have historically underperformed in specific markets are worth re-testing, and whether demo and sampling investments in those markets should be scaled up to capture the unlocked conversion. New item launches in 2026 and 2027 that hit those trade areas will perform against a different baseline than the launches that came before them.
The In-Store Partner Is Becoming More Capable, and Suppliers Should Match the Investment
Morris noted at EmTech that Walmart has the same number of associates today as it did when she joined six years ago, while revenue has grown substantially. The AI rollout is part of how Walmart makes that math work without expanding headcount, even as stores take on more complex roles in the company’s fulfillment network. The associate AI capability isn’t theoretical or imminent. Ask Sam is already operating at scale, VizPick has been running in apparel since 2021, the 44-language translation tool was announced in June 2025, and the EmTech remarks describe an expansion of literacy training across the entire workforce on top of capabilities that are already in production.
For 1P suppliers, the strategic read is that the in-store partner is becoming more capable, not less. The supplier organizations that benefit most will be the ones that recognize the shift early and rebuild their in-store investment thesis around it: what the field team does, what the retail services partner does, what the shelf intelligence vendor does, and how trade dollars flow against a fleet where the variance between best and worst stores is narrowing. The opportunity isn’t loud, and it won’t show up as a press release. It will show up in the numbers, for the suppliers who are looking.
Winning With Walmart
Winning With Walmart is an independent platform for suppliers, sellers, solution providers, and industry experts.
Articles are developed by staff editors, contributing experts, and trusted partners.
Some are researched and drafted with the assistance of AI tools and are reviewed, fact-checked, and edited by Winning With Walmart editors before publication.
Editorial judgment, sourcing decisions, and final approval rest with the publication's human editors in every case.
We are committed to accuracy and fairness. If you believe this article contains an error, we welcome your feedback.