AI-marketed products: 43% of Americans say they have been duped

What happened

Consumers are increasingly feeling misled by products marketed with artificial intelligence, according to new research from Auros, with 43% of Americans saying they have already been duped by an AI-advertised product.

The study surveyed 4,000 adults across the United States, the United Kingdom and Australia. Talker Research conducted the fieldwork, and the findings were shared in a press release that Retail Customer Experience reported. In each market, more than half of respondents said they do not trust AI-advertised products to be described accurately.

The research also finds that AI is becoming a more familiar part of shopping, and a majority of respondents plan to use it for some part of their holiday shopping this year. Bobby Meixner, vice president of marketing at Auros, described "a fascinating trust divide emerging around AI in commerce": shoppers are sceptical when AI is used to persuade them, but more receptive when they use it to make a better decision themselves.

Why it matters

For leaders in experience and digital transformation, the finding is less about AI's capability and more about where it sits in the customer journey. When a brand's AI-generated claims do not match the product a customer receives, the cost lands on trust in the brand itself. That makes accuracy of AI-produced product content an experience issue, not only a marketing or compliance one.

The same research points to a useful distinction. Consumers appear willing to accept AI as a decision aid that they control, but resist it when it is deployed to influence them without clear visibility. Organisations deploying AI across commerce and service will need to consider how they signal who or what is shaping a recommendation, and whether customers can verify it.

By the numbers

• 43% of Americans report they have already been duped by a product marketed with AI.
• 4,000 adults were polled across the U.S., U.K. and Australia.
• 59% of Australians, 58% of Britons and 53% of Americans say they do not trust AI-advertised products to be described accurately.

The Renascence take

The headline reads as a verdict on AI, but the data points to a narrower problem: an expectation gap between AI as a helper and AI as a salesperson. Most commentary will treat the trust deficit as a reason to slow down adoption. We read it as a design brief.

Consumers are not rejecting AI; they are rejecting ambiguity about whose interests it serves. Trust is asymmetric, so one overstated claim costs far more than a dozen accurate ones, and loss aversion will make that gap widen fast. A customer-obsessed operator should audit every AI-generated product claim against the actual product experience, and make it obvious when AI is advising the customer rather than selling to them.