Study reveals that Amazon and Walmart AI assistants identify fraud in the ‘Made in the USA’ seal

Research from Columbia Law School found that artificial intelligence assistants from Amazon (Alexa) and Walmart (Sparky) are capable of detecting products with false “Made in the USA” labels. However, retailers are not utilizing this ability to systematically remove misleading listings from their platforms.

The researchers suggest that the reasons for this technology not being widely used are commercial incentives, not technical limitations.

In a statement, Amazon said the idea that it was intentionally withholding information about the origin of products is incorrect.

Read also: Walmart reports 7.3% growth in revenue as consumers face inflation

A new survey details that artificial intelligence shopping assistants, employed by Amazon and Walmart, can identify several products falsely advertised as “Made in the USA” in their online stores. Despite this, companies do not implement this technology to combat misleading listings in an organized manner.

The study was released on July 31, 2026, by the Center for Law and Economics at Columbia Law School.

False ‘Made in the USA’ Product Claims at Retail Giants

The researchers investigated Amazon’s Alexa for Shopping and Walmart’s Sparky. The goal was to test the ability of these AI assistants to identify inconsistencies between product advertisements that explicitly stated they were “Made in the USA” and other information that indicated import or manufacturing abroad.

Research demonstrated that both chatbots could generally spot inconsistencies, highlighting that Amazon and Walmart have the technical ability to flag potentially misleading origin claims. However, the authors concluded that companies did not apply these capabilities comprehensively across all of their platforms.

The center, led by former US Federal Trade Commission (FTC) Chair Lina Khan, found that the restrictions were driven more by commercial interests than technical barriers. This raises questions about how retail companies are employing AI tools developed to assist consumers with their purchases.

Understand the case: Hamburger from Paraná unmasks customer scam that used AI to forge cockroach into snack

The report also indicated that misleading “Made in the USA” claims appear to be recurring on both platforms.

Artificial intelligence capabilities on the rise in retail

The study further mentioned that Alexa blocked certain queries related to “Made in the USA” products, but allowed similar queries for items labeled “Made in China.” According to the report, Alexa responded that this difference reflects a decision by the company aimed at protecting its international seller base.

Companies’ responses to the accusations

Walmart did not immediately respond to a request for comment.

Learn more: Fighter pays R$95 for two kiwis at Mercadão de SP; Internet users warn about ‘fruit scam’

Amazon said it continues to improve the way its AI shopping assistants provide information to customers, but reiterates that the suggestion of intentional country-of-origin data retention is false.

An Amazon spokesperson said: “We strive to provide customers with detailed information about the products in our store, including where they are made, whether that information is provided by manufacturers or third-party sellers. The suggestion that we intentionally withhold information about customers’ country of origin is incorrect.” The company representative added that “country of origin information, when available, is displayed on product detail pages, and we are continually working to improve Alexa for Shopping’s ability to accurately provide this information, making it even more accessible to customers.”

More on this story: Workplace ChatGPT starts copying individual writing styles

The company added that it takes action when sellers violate its marketplace policies and that the report’s broader conclusions about Amazon’s use of AI in shopping are misleading.

Implications and regulatory role of technology

The findings reignite broader debates about AI-powered shopping assistants, which retailers increasingly see as tools to personalize recommendations and boost consumer spending. The report argued that as these systems become more sophisticated, their implementation can continue to reflect commercial incentives, highlighting an essential role for policymakers and regulators. For consumers, this means that trust in product origin labels requires greater attention from authorities, ensuring that technological tools are used for protection and not just for profit.

FTC rules require that products advertised as “Made in the USA” be “wholly or virtually all” manufactured in the United States.

The previous year, the FTC had already asked Walmart and Amazon to step up their fight against “Made in the USA” claims made by third-party sellers, citing company policies that require sellers to provide accurate information.