Walmart AI is now the organizing principle behind the world’s largest retailer’s workforce plan. CEO Doug McMillon says artificial intelligence will “change literally every job,” yet the company aims to keep its global headcount roughly steady—around 2.1 million—for about three years. The pledge comes with a promise to help “everybody make it to the other side,” signaling a transition strategy that prioritizes training, redeployment, and new AI-enabled roles over mass layoffs. The approach sets a quantitative benchmark for how Big Retail intends to absorb AI-driven productivity gains.

Key Takeaways

– shows Walmart will keep global headcount near 2.1 million for roughly three years while redeploying staff into AI-augmented roles that expand. – reveals on May 21, 2025, a restructuring cut about 1,500 roles in tech, e-commerce fulfillment, and advertising, while opening new positions. – demonstrates the U.S. accounts for about 1.6 million of Walmart’s 2.1 million workers, magnifying domestic impact from AI-driven job redesign. – indicates leaders said on Sept. 26, 2025, AI will change “literally every job,” with internal “agent” chatbots and “agent builder” roles scaling. – suggests July 2025 hiring of Daniel Danker and multi-year training underpin a flat 2.1 million headcount goal despite productivity gains and role shifts.

Walmart AI strategy: keeping 2.1 million employees flat

Walmart’s plan is deceptively simple: hold total employment steady even as AI rewires how work gets done. Executives now target a roughly flat headcount—about 2.1 million globally—for the next three years, a timeline that gives the company room to scale automation while retraining associates into roles where human judgment, customer service, and oversight of AI “agents” are rising in value [2].

Flat does not mean static. Management expects certain job families to shrink as others grow. That implies a high rate of internal mobility and a tighter focus on matching skills to changing demand. In practice, that means designing pathways for associates whose tasks become automated to move into higher-impact work, supported by training, coaching, and new role archetypes that combine AI fluency with domain expertise.

The plan also signals a measured pace. Rather than chasing headline headcount reductions, Walmart appears to be normalizing AI-driven efficiency within a stable employment envelope, absorbing gains through mix changes. The message to workers is clear: the organization will change, but the company aims to keep the overall job count near its current level while people reskill.

What Walmart AI means for job design: “every job” changes

At a Bentonville workforce forum, McMillon’s line—AI will change “literally every job”—was less a warning than a blueprint. In frontline settings, AI can streamline repetitive tasks and surface recommendations; in corporate roles, it can accelerate analysis, planning, and creative iteration. Walmart has begun building internal AI “agents” to execute routine steps and proposes new “agent builder” roles to create, maintain, and govern these tools across the enterprise.

This shift reframes job scopes. Instead of replacing associates one-to-one, AI embeds into workflows, changing the ratio of time spent on manual steps versus decision-making. For example, a support agent might oversee multiple AI-driven processes, intervening when the system flags exceptions. A merchandiser might spend less time collecting data and more time setting strategy and validating AI-generated options.

The practical effect is a sweeping reskill-and-redesign agenda. Teams will need policies on when to trust, edit, or override AI outputs; managers will need metrics for agent performance; and associates will need access to training that maps to new task mixes. The “every job” statement is a statistical forecast of task composition, not a prediction of mass displacement [2].

Reconciling May layoffs with a flat three-year headcount

In May, Walmart announced a restructuring that eliminated roughly 1,500 roles across global technology, e-commerce fulfillment, and advertising. The company said the moves were part of a simplification push and came alongside openings in other areas—an early example of how reductions in some functions can be offset by growth elsewhere [3].

Seen alongside the three-year flat headcount target, May’s cuts look like a recalibration rather than a reversal: prune duplicative or lower-priority roles, then redeploy hiring toward capabilities that complement AI. For employees, the takeaway is that churn will be localized—specific teams may shrink or move—while the broader enterprise manages to a stable headcount. For investors and labor economists, it suggests Walmart expects productivity gains to flow into service quality, speed, and new capabilities rather than immediate large-scale staff reductions.

Training and transitions: making “everybody” cross the AI chasm

McMillon’s pledge to help “everybody” get to the other side puts training at the center of execution. Chief People Officer Donna Morris underscored that the company must “do our homework” to determine the new ratios of jobs in an AI-powered operation—how many associates in which roles, supported by how many AI agents, over what time horizon [4].

Quantitatively, that means building models for task time redistribution, forecasting which tasks AI will compress by 10%, 30%, or more, and then translating those deltas into revised job descriptions. It also means budgeting training hours per associate, setting pass rates for new certifications, and tracking internal mobility rates as a leading indicator of a healthy transition.

Expect Walmart to focus on stackable credentials—short courses that culminate in role-ready proficiency—and to set adoption KPIs for AI tools: agent utilization rates, exception rates requiring human intervention, and time-to-proficiency for new hybrid roles. The enterprise may also align incentives to prioritize redeployment over external hiring where skills can be transferred with targeted upskilling [4].

Inside the build: agents, agent builders, and the July hire

Getting from vision to code requires product leadership. In July, Walmart hired Daniel Danker, a veteran product executive, a move consistent with building internal “agent” platforms that scale across stores, supply chain, merchandising, and customer service. The leadership emphasis pairs with McMillon’s line about creating opportunity for “everybody,” signaling investment in tooling that amplifies human work rather than just automates it away [1].

On the ground, agent platforms could standardize best practices, embed compliance checks, and pre-assemble workflows associates can customize. “Agent builders” would act as internal product managers and prompt engineers, curating datasets, designing guardrails, and ensuring outputs meet operational and ethical standards. As those roles crystallize, they create visible landing spots for associates moving out of tasks most affected by automation [1].

Baseline workforce math and what to watch

Walmart employs about 2.1 million people worldwide, including roughly 1.6 million in the United States. Any AI-induced role mix changes will therefore land most heavily on U.S. operations by absolute numbers, even if percentages are similar globally. Executives have been explicit that composition will change as adoption expands, even as the total count remains steady over the three-year horizon [5].

What should stakeholders monitor? First, vacancy postings by function—are “agent builder,” data, and operations-adjacent roles growing as others recede? Second, internal mobility and retraining metrics—what share of associates move into AI-augmented roles within 6–12 months? Third, service and productivity KPIs—does order accuracy, shelf availability, or pickup speed improve alongside stable staffing levels?

Governance is another watchpoint. As AI agents scale, the company will need clear escalation thresholds, audit trails, and bias-monitoring protocols. Workforce acceptance is just as critical: survey data on associate trust in AI guidance, coupled with incident reporting, can predict adoption risk. If the data shows rising productivity per associate and stable or improving customer outcomes, the flat headcount strategy will look prescient.

Finally, the timeline matters. A three-year window allows Walmart to iterate on AI deployments, integrate feedback, and phase training without overwhelming operations. It also offers time to test wage and scheduling policies that reflect new task mixes. If Walmart can demonstrate net skill upgrades and internal promotion pathways at scale, its “everybody to the other side” promise will read not just as rhetoric, but as a measurable outcome [5].

Sources:

[1] Yahoo Finance (Fortune syndicated) – Walmart CEO wants ‘everybody to make it to the other side’ and the retail giant will keep headcount flat for now even as AI changes every job: https://finance.yahoo.com/news/walmart-ceo-wants-everybody-other-222445979.html

[2] The Wall Street Journal – Walmart CEO Issues Wake-Up Call: ‘AI Is Going to Change Literally Every Job’: https://www.wsj.com/tech/ai/walmart-ceo-doug-mcmillon-ai-job-losses-dbaca3aa [3] CNBC – Walmart plans job cuts in restructuring push to simplify operations: www.cnbc.com/2025/05/21/walmart-plans-job-cuts-in-restructuring-push-to-simplify-operations.html” target=”_blank” rel=”nofollow noopener noreferrer”>https://www.cnbc.com/2025/05/21/walmart-plans-job-cuts-in-restructuring-push-to-simplify-operations.html

[4] PYMNTS – Walmart CEO Foresees AI Transforming ‘Literally Every Job’: www.pymnts.com/walmart/2025/walmart-ceo-foresees-ai-transforming-literally-every-job/” target=”_blank” rel=”nofollow noopener noreferrer”>https://www.pymnts.com/walmart/2025/walmart-ceo-foresees-ai-transforming-literally-every-job/ [5] Reuters – Walmart to cut about 1,500 roles, memo says: www.reuters.com/technology/walmart-plans-cut-about-1500-roles-memo-2025-05-21/” target=”_blank” rel=”nofollow noopener noreferrer”>https://www.reuters.com/technology/walmart-plans-cut-about-1500-roles-memo-2025-05-21/

Image generated by DALL-E 3