What happened
Walmart CEO John Furner announced the launch of the Walmart Agent Gateway and Agent Profiles at the Masters of Scale Summit on October 21. The Agent Gateway is a server built on the Model Context Protocol (MCP) that allows approved third-party AI agents, including ChatGPT, Gemini, and Meta’s Muse, to connect to Walmart and Sam’s Club inventory to search for products and complete purchases. Starting next month, Agent Profiles will create distinct profiles for each within a customer’s Walmart account. These profiles allow customers to separate agent-made purchases from their own, set spending limits, or require approval for every transaction. Walmart has established strict terms for these agents, prohibiting access to payment card data and limiting data usage to what is necessary for the task. This move positions Walmart as open to agentic commerce, contrasting with Amazon, which recently blocked Meta’s Muse from accessing its platform due to security concerns.
Walmart CEO John Furner announced the introduction of the Walmart Agent Gateway and Agent Profiles at the Masters of Scale Summit in San Francisco on October 21. The Agent Gateway is a server-based tool built on the Model Context Protocol (MCP) that facilitates connections between approved third-party AI agents and Walmart’s retail infrastructure. According to the report, this gateway allows agents such as ChatGPT, Gemini, and Meta’s Muse to search Walmart and Sam’s Club inventory and execute purchases on behalf of customers. The system is designed to allow customers to have agents reorder usual brands without the agent needing access to the customer’s personal information or full purchase history.
Starting next month, Walmart will implement Agent Profiles, which create a distinct profile for each within the customer’s existing Walmart account. These profiles are automatically created the first time an agent is used through the gateway. Purchases made by an agent are displayed separately from the customer’s own transactions. Customers can manage these profiles by requiring approval for every transaction or setting specific spending limits. Walmart describes these controls as similar to adding an authorized user on a credit card, and profiles can be deleted at any time. The service does not require a Walmart+ membership.
Walmart has established specific terms of service for AI agents using the gateway. Agents are prohibited from accessing payment card data and are restricted to using only the information necessary for the specific task. They cannot act beyond the shopper’s intended scope, and non-compliant agents will be blocked. The standard Walmart returns process applies to agent-made purchases, even if the agent did not act as intended. Agents see the same prices as human shoppers, with all offers and fees displayed before the purchase is completed.
The announcement highlights a strategic divergence from Amazon, which cut off Meta’s Muse from Amazon.com on September 20. Amazon cited that Meta had not notified it of Muse’s access, that the agent did not identify itself while browsing, and that unauthorized agents pose privacy and account-security risks. Meta responded that Muse has no visibility into users’ passwords or payment details. Walmart’s approach explicitly addresses these concerns through the MCP-based gateway and profile isolation.
Why it matters
This launch represents a significant shift in e-commerce infrastructure, moving from human-only browsing to standardized, secure interfaces for autonomous AI agents. By adopting the Model Context Protocol, Walmart is aligning with emerging industry standards for AI interoperability, potentially setting a precedent for how major retailers integrate with third-party AI ecosystems. The introduction of Agent Profiles addresses key privacy and security concerns by isolating agent actions from the primary user account, allowing for granular control over spending and data access. This development is critical for the broader adoption of agentic commerce, as it provides a framework for trust and accountability that may encourage other retailers to follow suit, fundamentally changing how consumers interact with online shopping platforms through AI intermediaries.
The launch of the Agent Gateway marks a concrete step toward the standardization of agentic commerce. By utilizing the Model Context Protocol, Walmart is adopting a technical standard that allows for interoperability between different AI models and retail systems. This move suggests that the industry is moving away from proprietary, closed-loop AI shopping experiences toward open, protocol-driven integrations that can support a wider range of third-party agents.
The introduction of Agent Profiles directly addresses the security and privacy concerns that have hindered the adoption of autonomous shopping agents. By isolating agent actions into separate profiles with granular control over spending and approval, Walmart provides a mechanism for consumers to trust AI intermediaries with their purchasing power. This structural change is significant because it shifts the responsibility for security from the itself to the retail platform’s infrastructure, potentially reducing the risk of unauthorized transactions or data leaks.
This development has practical implications for the competitive landscape of e-commerce. By positioning itself as open to agentic commerce while Amazon blocks such agents, Walmart may attract users who prefer the convenience of AI-driven shopping. The ability to use agents like ChatGPT and Gemini to shop at Walmart could drive increased traffic and sales through these AI interfaces, creating a new channel for customer acquisition and retention.
Interactive Mechanism: How It Actually Works
Explore the underlying technology behind this development interactively.
crm_get_transaction(id='4092').An agent must create a draft calendar event for Tuesday at 2 p.m. Which evidence would establish the requested result?
What to watch next
Monitor the specific list of approved AI agents beyond ChatGPT, Gemini, and Muse, and whether other major retailers like Amazon or Target announce similar gateway solutions. Watch for consumer adoption rates of Agent Profiles and any reported security incidents or policy adjustments in the early months of deployment. Additionally, observe how the Model Context Protocol standard evolves and whether it becomes the dominant technical standard for retail AI integration.
The specific timeline for the full rollout of Agent Profiles and the initial list of approved AI agents beyond those named (ChatGPT, Gemini, Muse) will be key indicators of the program’s scope. Monitoring whether other major AI developers seek approval for their agents will reveal the breadth of the ecosystem Walmart is building.
Consumer feedback and adoption metrics for Agent Profiles will be crucial. If users find the approval and spending limit controls cumbersome, it could limit the utility of the feature. Conversely, high adoption rates would validate the model for other retailers to emulate.
The reaction from Amazon and other major retailers will be significant. If Amazon maintains its blocking stance, it could create a fragmented market where agentic commerce is only available on certain platforms. If other retailers follow Walmart’s lead, it could accelerate the of AI-driven shopping across the industry.