AI-Generated Ads and Consumer Trust: 53% of U.S. Shoppers Doubt Product Claims

AI can make more advertising, faster. But new consumer research suggests that production efficiency does not automatically create buyer confidence. In a survey released October 16, 2026, 53% of U.S. adults said they do not trust products advertised with AI to be described accurately, while 43% said they had felt duped by a product advertised with AI. For marketing leaders, the issue is no longer just whether a team can generate a polished image or headline. It is whether the creative makes the real product easier to understand, or harder to trust.

The study was commissioned by Auros and conducted by Talker Research among 4,000 adults in the United States, United Kingdom and Australia. The U.S. sample included 2,000 adults surveyed from August 26 to September 1, 2026. Those findings are a warning signal, not proof that AI itself caused distrust: they reflect respondents’ perceptions in a company-commissioned survey, not controlled experiments tracking purchase behavior. Auros and Talker Research’s full release also reports that 44% of Americans go out of their way to do business with brands that do not use AI in advertising.

The real risk is an expectation gap

The numbers point to a specific commercial problem: consumers fear that the ad will overpromise what arrives. The 53% who distrust accuracy are reacting to the product representation, while the 43% who say they have felt duped describe a credibility failure after exposure. A synthetic visual that exaggerates size, texture, color, results or context can make a campaign more attention-grabbing and still increase doubt about the underlying offer.

That distinction matters because “AI-generated” covers very different practices. It can mean editing a background, drafting copy, producing a wholly synthetic model, or depicting a product that does not exist in the form shown. Treating all of these as one category obscures where the brand risk sits. The strongest operational question is not simply whether AI touched an asset. It is whether a reasonable customer could infer a product feature, result or endorsement that the business cannot substantiate.

AI assistance and AI advertising are different propositions

The Auros release frames consumers as wary of AI retail ads but open to AI as a shopping tool. That is a useful distinction for strategy. A shopper may welcome an assistant that compares specifications, finds a price or answers a question, while remaining skeptical of a persuasive image that looks too perfect or does not match the item delivered.

For marketers, this means AI’s value may be easier to earn in utility than in spectacle. Use it to help customers navigate choices, summarize accurate product information or adapt an approved message to a channel. If generative creative is used, maintain a clear connection to the real product, its verified attributes and the actual customer experience. Disclosure can clarify process, but it cannot repair a misleading depiction or weak product evidence.

Build a product-truth review into the creative workflow

Before scaling AI-assisted campaigns, teams should add a lightweight review that checks the claim and the image together. Reviewers should compare generated assets against real product photography, packaging, specifications, pricing, availability and any performance claim. A technically accurate caption cannot rescue a visual that creates a false impression, and a photorealistic image should not be treated as evidence that the depicted feature is real.

Then test what buyers actually understand. Run controlled versions of a campaign with real photography, AI-assisted variations and clear explanation of the product. Track more than click-through rate: include product-page engagement, add-to-cart, conversion, returns, customer-service questions and complaints about mismatch. Where possible, ask a short post-exposure question about what the customer believes the item includes. If a generated version wins attention but raises confusion or returns, the apparent media efficiency may be disguising a downstream cost.

Turn AI speed into a trust advantage

These results do not mean that brands should stop using AI. They do mean that AI adoption is visible to consumers through the finished experience, and that audiences may judge the product claim more harshly when the execution feels synthetic or implausible. The study is commissioned research, not an industry-wide causal benchmark, so each company should validate the signal with its own customers. Still, the scale of the U.S. responses makes “AI-assisted” a poor substitute for a real quality-control standard.

Marketing leaders should set explicit boundaries for what AI can invent, require human approval for product claims and visual fidelity, and preserve a record of the source assets and evidence behind each campaign. The goal is not to conceal useful technology or to add process for its own sake. It is to make faster creative production compatible with accurate promises.

Real Internet Sales helps businesses build measurable, AI-enabled marketing systems without losing sight of the customer experience. Call 803-708-5514 or visit realinternetsales.com to discuss your strategy.