Study: Amazon and Walmart AIs Spot Fake 'Made in USA' Labels — But Retailers Aren't Alerting Shoppers

A new study from Columbia Law School's Center for Law and the Economy finds that both Amazon's Alexa For Shopping and Walmart's Sparky chatbot can detect misleading 'Made in USA' claims — yet the retailers are not using those detections to warn buyers or remove deceptive listings, Reuters reports.

What the Researchers Did

Researchers tested the two shopping assistants directly and presented product pages where a bold 'Made in USA' label conflicted with country-of-origin information buried elsewhere on the same listing. Both chatbots reliably identified these conflicts, showing the AI systems have the technical capacity to spot such mismatches.

Walmart

What the Bots Said

When researchers asked why the platforms weren't acting on the detected mismatches, neither assistant cited technical limitations. According to the study, Amazon's Alexa acknowledged that these false claims can harm U.S. brands but implied the company tolerates the issue until it becomes a direct cost to Amazon. Walmart's Sparky was more blunt, characterizing the lack of action as a business calculation rather than a legal or technical constraint. The study also notes a curious inconsistency: Alexa will answer questions about products made in China but blocks or refuses similar queries about 'Made in USA' claims.

Regulatory Context

The Federal Trade Commission requires that anything labeled 'Made in USA' be "all or virtually all" made domestically. The FTC previously urged both retailers to address false country-of-origin claims by third-party sellers. The Columbia center that produced this report was founded by former FTC Chair Lina Khan, who while at the agency separately sued Amazon over alleged monopoly practices — a case that remains active.

Responses

Amazon told Reuters that origin information appears on product pages when available and that the company is working to make that information easier to find. Walmart declined to comment to Reuters. Lina Khan and the study's authors argue the findings underscore a need for stronger regulatory oversight so consumers and domestic manufacturers are protected as AI tools grow more capable.

Bottom line: The AIs can spot inconsistent origin claims, but current platform choices — not technical limits — are preventing that capability from protecting shoppers and honest sellers.