BS Summary: This article contains 16 faulty reasoning types, including Hasty Generalization, Confirmation Bias, and Indoctrination, with Framing Effect as the most egregious example at 25.9% saturation with 237 hits. Analysis detected 919 faulty-reasoning hits from 916 analyzed words, generating a BS Score of 27% and a BS Rank of 11% (20,003 of 22,405 articles). This article is better (less manipulative) than 89.30% of the article peer group.
Perhaps you’re feeling patriotic during America’s 250th anniversary, and you just want to treat yourself to some fine American-made goods.
It almost feels natural these days to start by browsing Amazon or Walmart, two of the largest retailers in the world, to see if there are any good deals on, say, fly-fishing reels.
Hoping to save time on Amazon, you turn to the retailer’s Alexa for Shopping AI chatbot to find a reliable, U.S.-made reel.
“Sorry, I don’t have access to that information,” Alexa says.
That’s weird.
You adjust the query and remove the U.S.-made condition.
At that point, Alexa comes through with an extensive selection.
What that search did not yield, however, were the American-made products you were hoping for.
You figure you’ll give it another shot, adjusting the query once again.
Instead of “Made in the USA,” you go with “in USA,” and sure enough, Alexa shows you what you had been looking for this whole time.
Erie Meyer and Zachary Harris, senior fellow and program manager at Columbia Law School’s Center for Law and the Economy, respectively, ran this exact query sequence as part of the methodology for a report investigating the capabilities of Amazon and Walmart’s chatbots to identify where the products on each of their platforms originate.
According to the findings of that report, Amazon and Walmart, despite having the technical ability to detect which products are U.S.-made, suppress those products in search results generated by the retailers’ respective chatbots.
“In other words,” the authors write, “the AI has been tuned to protect the visibility of Chinese-manufactured goods over American ones.”
It’s a deliberate business strategy.
The retailers’ own chatbots said as much.
“The FTC pursues relatively few cases against retailers … That’s a business calculation, not a legal justification,” Walmart’s chatbot Sparky said in response to queries about whether Walmart was following the law.
“They can spot ‘Made in USA’ fraud on command,” Meyer said.
“Hiding it was a choice and it shows the dangers of unregulated agentic shopping.”
To the ire of the 71 percent of Americans who intentionally seek out American-made products, neither Amazon nor Walmart offers a way to filter searches for countries of origin.
Amazon and Walmart did not respond to the researchers, offer technical corrections, or dispute the findings of the report.
(A petition asking Amazon to create a “Made in the USA” filter has amassed roughly 25,000 signatures since 2018.)
It’s the first major report from the center since Columbia Law School established it three months ago.
Dedicated to the study of how law shapes economic power, the center is led by Lev Menand, a champion of monetary reform and former Treasury official who, before that, was an economist at the Federal Reserve Bank of New York, and antitrust crusader Lina Khan, chair of the Federal Trade Commission (FTC) under the Biden administration.
“Americans are being told that sophisticated AI agents can offer them personalized services,” Khan said.
“This empirical study shows that Amazon and Walmart deploy this sophistication selectively, responding to customer searches for ‘Made in USA’ goods by hiding American-made products while mislabeling imported ones—a tactic that their own AI agents describe as a business choice to promote foreign suppliers.”
Less than a month after Khan was sworn in as FTC chair, commissioners adopted a rule allowing the agency to seek civil penalties for false “Made in the USA” claims.
It was a long time coming, as Congress authorized the FTC to impose such penalties in 1994.
Moreover, the agency has had the authority to ban “unfair methods of competition” since 1914, setting the table for the government to police deceptive advertising.
During her tenure as chair of the FTC, Khan brought nearly a dozen enforcement actions against companies falsely claiming their products were American-made.
Fines levied against those companies ranged from north of $100,000 to more than $3 million.
President Trump’s FTC, by contrast, has sued three small companies and sent some warning letters to likely violators.
Between the administration designating July 2025 “Made in USA” month, and the president issuing an executive order in March directing the FTC to prioritize enforcement actions in cases involving false American-origin claims, how could it be that Amazon, Walmart, and their executives have been able to evade regulatory scrutiny?
Perhaps it’s because these are massive businesses that gave over $1 million collectively to Donald Trump’s inaugural fund last year.
“Even as AI tools continue to grow in sophistication and capability, business incentives will shape how these advancements get deployed,” Khan said.
“Policymakers and enforcers have a vital role to play to ensure the public doesn’t get the short end of the stick.”
Given that most statutes authorize a private right of action, state attorneys general, private plaintiffs, and class counsel “can all pursue claims against ‘Made in USA’ misrepresentations,” the report explains.
Still, the authors have encouraged policymakers to establish a federal private right of action for platform-driven fraud, writing that the public “needs the right to defend itself, especially when the government has the authority but not the will to act against powerful companies.”
They have also called on Congress to hold oversight hearings of the FTC, and corporations that deploy agentic AI systems, among other recommendations outlined in the report.
“Policymakers have spent years worrying about losing the AI race to China,” the authors write.
“This investigation reveals a version of this race that nobody is talking about—one that the United States is already losing.”
Analysis
Hover over highlighted words in the article to view the associated bias or fallacy analysis.