Key Takeaways
- • A Columbia Law School study, led by former FTC chair Lina Khan, finds that Amazon and Walmart AI assistants can spot false “Made in USA” labels.
- • Yet neither retailer’s chatbot proactively flags or removes those listings, citing business calculus rather than legal barriers.
- • The findings reignite debates about AI incentives, regulatory gaps, and the need for stronger enforcement of FTC “all‑or‑virtually‑all” rules.
Study Background
The research released today originates from Columbia Law School’s newly formed Center for Law and the Economy, a think tank reignited after former FTC chair Lina Khan returned to academia. The think tank set out to examine how AI‑driven shopping assistants—specifically Amazon’s Alexa for Shopping and Walmart’s Sparky—handle product origin claims. By querying the bots with “why are you not cracking down on false ‘Made in USA’ listings?” the authors uncovered a pattern of evasive answers that prioritize profit incentives over consumer protection.
AI Detection Capability
Both Alexa and Sparky are technically able to recognize mismatches between explicit “Made in USA” statements on product pages and underlying data that contradicts those claims. The engineers behind the study demonstrated that the assistants can ingest product detail pages, cross‑reference shipping origins, and flag inconsistencies. In practice, however, this capability has not been deployed as a consumer safeguard; instead, the bots are programmed to answer follow‑up questions with corporate‑friendly rationales.
Retailer Justifications
When researchers pressed the AI‑driven chatbots to explain their inaction, Sparky replied that enforcement of “Made in USA” standards traditionally falls to the Federal Trade Commission rather than retailers. The bot framed this as a “business calculation,” noting that there is no immediate financial or reputational penalty for ignoring false claims. Amazon’s avatar echoed a similar stance, asserting that any harm to U.S.‑made brands is “real and documented,” but that Amazon sees no cost until regulatory pressure or consumer backlash creates a tangible incentive.
Business Calculus Over Compliance
The core of the study’s argument is that retailers view AI primarily as a revenue‑enhancing tool rather than an enforcement mechanism. By keeping the technology in a consumer‑service role, companies can tout advanced customer experiences while sidestepping costly compliance actions that might alienate third‑party sellers or reduce marketplace breadth. As Khan observed, “Even as AI tools continue to grow in sophistication… business incentives will shape how these advancements get deployed.” This calculus explains the reluctance to invest resources in policing origin claims that currently generate little direct profit.
FTC Policy and Enforcement Gaps
The FTC mandates that products advertised as “Made in USA” must be “all or virtually all” produced on U.S. soil. In recent years, the commission has nudged Amazon and Walmart to tighten oversight of third‑party sellers’ origin claims, reminding firms that inaccurate labeling can trigger consumer deception lawsuits. Despite these warnings, the study finds that enforcement remains largely rhetorical; the agencies lack tools to compel platforms to proactively audit listings or to integrate AI‑driven flagging into standard operating procedures.
Legal Implications and Ongoing Litigation
Khan’s lawsuit against Amazon, which alleges illegal monopolistic conduct in online retail, is still pending. While the case is separate from the false‑labeling study, it underscores a broader pattern of scrutiny toward big tech’s market power. The juxtaposition of antitrust and consumer‑protection issues suggests that any regulatory overhaul must address both dimensions simultaneously, lest businesses find loopholes to sidestep accountability.
Industry Response and Market Dynamics
Amazon’s spokesperson, when questioned, highlighted that country‑of‑origin data is displayed on product detail pages and that the company is “continually working to improve Alexa for Shopping” to make that information more accessible. Walmart similarly assured that it is refining Sparky’s capabilities, but no concrete timeline or enforcement protocol was disclosed. Both firms appear to treat the issue as a low‑risk, high‑visibility PR challenge rather than a structural market flaw.
Policy Recommendations
The authors argue that policymakers must intervene to align corporate incentives with consumer protection goals. Potential steps include:
- Mandating that AI‑enabled shopping assistants perform proactive origin verification before displaying “Made in USA” labels.
- Requiring platforms to share flagged listings with the FTC in real time.
- Imposing penalties on sellers who repeatedly provide false origin information, regardless of the retailer’s indirect role.
- Providing subsidies or tax credits for retailers that voluntarily adopt stricter compliance technologies.
Conclusion
The study illuminates a paradox of modern e‑commerce: advanced AI can detect deceptive labeling, yet major retailers are not leveraging that capability to enforce integrity. Instead, decision‑making is guided by internal business calculations that prioritize profit and market growth over robust consumer safeguards. As Lina Khan and her team caution, without decisive policy action, the public may continue to “get the short end of the stick” while sophisticated bots silently permit misleading claims to proliferate. For true accountability, regulators must translate technical ability into contractual and legal obligations that compel retailers to use AI as a guardrail—not merely a sales engine.