American Express has unveiled a new technical framework designed to help card issuers verify purchases made by artificial intelligence-powered shopping agents, marking a significant step forward in the rapidly evolving landscape of automated e-commerce.

The payments giant released its Intent Intelligence framework, detailing a standardized method for recording a shopper’s specific purchase instructions before autonomous AI agents execute transactions on their behalf. As consumers increasingly experiment with digital assistants capable of browsing products, comparing prices, and completing checkouts, financial institutions face mounting challenges in authenticating these non-human transactions and managing the associated risks.

Under the newly published system, card issuers can utilize authenticated records of customer intent to help detect fraudulent activity, streamline payment approvals, and swiftly resolve disputes when transactions do not go as planned. For retailers, the framework offers a vital mechanism to determine whether an ordering mistake originated from an AI agent misinterpreting a command or from a fulfillment error on the merchant’s end.

While American Express confirmed that Intent Intelligence does not yet feature live integrations, the publication of the framework is intended to encourage other payment networks and technology providers to adopt a collaborative approach rather than introduce proprietary capabilities immediately.

"Providing authorized purchase intent before an agent transacts can give issuers greater confidence that an agent is acting as a customer intended, while also helping issuers proactively manage risk and reduce the chance of costly disputes across the payments ecosystem," said Luke Gebb, head of global innovation at American Express, in the company’s official announcement.

The rollout arrives as financial institutions race to develop robust infrastructure for agentic commerce. These emerging tools allow autonomous software to navigate online storefronts and execute purchases independently. Yet, despite the technological enthusiasm, everyday consumers remain cautious about relinquishing complete financial control to algorithms.

Recent industry research underscores this consumer hesitation. In a survey of 1,000 U.S. adults released in August by payments company NMI, 58% of respondents expressed worry that an AI shopping agent might mistakenly purchase the wrong item, size, color, or brand. Furthermore, 60% of surveyed consumers voiced concerns that an autonomous agent could easily overspend or exceed their intended budget.

Although American Express initially introduced Intent Intelligence earlier in the year, the latest release makes its detailed technical architecture accessible for external companies to build upon. The overarching goal, according to the company, is to help advance a common, industry-wide standard for handling consumer intent within agentic commerce.

What the New American Express Framework Means for Retailers

For online merchants and traditional retailers alike, purchasing errors driven by artificial intelligence introduce a complex operational challenge. Historically, if a merchant fulfilled an order exactly as transmitted through checkout systems, the transaction was considered valid. In an era of autonomous shopping agents, however, a merchant could process an order flawlessly only to face a contentious chargeback dispute because the AI agent bought an item the human shopper never actually requested.

This is precisely where documented purchase instructions—or verified purchase intent—become essential. By securely capturing and preserving what the customer originally approved, the framework provides an audit trail that establishes whether the AI agent faithfully executed the user’s explicit commands.

Intent Intelligence is one of several specialized services originally introduced through the American Express Agentic Commerce Experiences (ACE) developer kit. The company emphasizes that this framework differs fundamentally from current implementations by transmitting a customer’s authorized purchase intent to the card issuer much earlier in the transaction lifecycle, well before the agent executes the final payment.

The framework supports a wide variety of advanced purchase controls designed to give consumers granular oversight. Shoppers can establish strict maximum spending limits, set expiration dates for specific purchasing tasks, and designate preferred merchants. Additionally, consumers can specify precise timing conditions for execution, such as instructing an agent to purchase an item only when it returns to stock or waiting to complete transactions during major retail events like Cyber Monday.

Beyond basic product acquisitions, the system is built to handle multi-layered and complex shopping scenarios. For instance, a consumer could authorize an AI agent to purchase optional flight insurance only if it successfully secures the accompanying airline tickets. The architecture also accommodates multi-merchant transactions, allowing a single shopping session to seamlessly encompass buying shoes from one digital storefront and matching socks from another.

How American Express’s Intent Intelligence Framework Works

The underlying architecture of Intent Intelligence is designed to operate seamlessly across diverse agentic commerce ecosystems, regardless of which platform or entity collects the initial shopper instructions. Depending on how a particular AI agent manages payment flows, it can transmit the authorized instructions directly to a card issuer, or route them through a digital wallet provider or another authorized financial intermediary.

By establishing a flexible pipeline for intent data, American Express aims to ensure that transaction verification remains consistent even as consumers use a wide variety of software applications to manage their day-to-day shopping needs.

Why AI Shopping Errors Create New Payment Risks

In addition to assisting card issuers with purchase verification, the Intent Intelligence framework is closely tied to broader consumer and merchant protection initiatives outlined by American Express.

When the company originally rolled out its ACE developer kit, it revealed plans for a specialized coverage tier known as Agent Purchase Protection. This proposed protection is intended to cover eligible transactions involving registered AI agents that utilize the developer kit and successfully transmit authenticated purchase instructions. For merchants, the primary advantage is a significant reduction in costly chargebacks and complex dispute resolution processes.

However, American Express has confirmed that these protections are not yet active, as there are currently no live ACE integrations utilizing the Intent Intelligence framework required to qualify transactions. In the interim, card members continue to rely on the standard consumer protections and benefits already associated with their eligible American Express cards.

This evolving landscape of automated purchasing also introduces shifting dynamics regarding accountability when things go wrong. In the NMI consumer survey, 34% of respondents stated they would hold the retailer directly responsible if an AI tool executed a purchase prematurely. Meanwhile, 24% indicated they would blame their payment provider, and 62% reported that they would hold the AI software provider primarily accountable for any errors.

Where American Express Fits in Agentic Commerce

American Express is far from alone in crafting security frameworks and protocols to govern AI-powered retail. The entire financial technology sector is moving rapidly to establish standards for automated commerce.

In the spring of last year, major card networks Visa and Mastercard both debuted dedicated offerings allowing developers to integrate AI agents directly into their established payment infrastructures. Visa’s Intelligent Commerce program provides developers with application programming interfaces and technical tools to embed payment functions smoothly into conversational and autonomous AI agents, backed by partnerships with major technology firms including OpenAI, Microsoft, Anthropic, Stripe, and Samsung.

Mastercard introduced Agent Pay, a technology designed to allow AI agents to handle transactions on behalf of users across both consumer-facing applications and business-to-business commerce environments.

Concurrently, major technology and software companies have been pioneering their own protocol standards. OpenAI and Stripe co-developed the Agentic Commerce Protocol to facilitate secure connections between AI agents and merchant checkouts. Google launched its Agent Payments Protocol, which relies on digitally signed records to document user authorizations, quickly securing early adoption from major enterprise platforms like Salesforce, Shopify, Cloudflare, and Etsy.

Google expanded its ecosystem further with the Universal Commerce Protocol, an open standard developed alongside Shopify and other industry partners to enable frictionless checkout directly inside AI-driven shopping environments. Prominent retailers such as Newegg, Wayfair, Walmart, and Macy’s have since adopted these protocols, allowing consumers to complete purchases natively within advanced artificial intelligence interfaces.