Consumer Survey Puts Trust in AI Advertisers Under the Spotlight

A survey of 446 U.S. adults found that the timing of an AI advertising disclosure was tied to how much people trusted the advertiser afterward. Consumers who learned about the AI use before buying responded more favorably than those who discovered it later.

An online ad can persuade someone to buy an unplanned product, and the customer can end up satisfied with that purchase. Even so, the advertiser may face a trust problem if the customer later learns that artificial intelligence created the ad and that the ad did not disclose it at the time.

That is the central finding from a ContentEngine survey comparing two groups of 223 U.S. adults. One group saw a scenario in which the ad clearly disclosed before the purchase that AI created its words and images. The other group learned about the AI involvement only after buying the product and using it successfully.

The difference showed up in the measures brands care about most. People who learned about the AI use afterward were less likely to trust the advertiser, buy from it again or overlook the lack of disclosure. The result gives advertisers a straightforward practical warning: if AI use is material to how consumers understand an ad, disclosure belongs before the decision to act.

The survey gave every respondent the same favorable outcome. The advertised product was something the person had not planned to buy, the ad persuaded the person to purchase it, and the person was satisfied after using it. The timing of the AI information was the key feature separating the scenarios.

The responses produced three clear differences:

The gaps were roughly 12 to 13 percentage points. After adjustment for the survey’s eight comparisons, the differences in trust, repeat purchase intent and feeling misled remained statistically significant. The reported 95 percent confidence intervals also stayed away from zero for those measures.

The finding matters because product satisfaction was held constant in both scenarios. A customer who liked the product still reacted less favorably when the AI involvement surfaced after the purchase. Immediate sales and customer satisfaction therefore may leave out an important part of the advertising result: how the audience interprets the brand’s conduct later.

A ContentEngine survey of 446 U.S. adults found that the timing of an AI advertising disclosure was associated with consumer trust, repeat-purchase intent and feelings of being misled. Respondents told about AI before an unplanned purchase responded more favorably than those who learned about it after purchasing and successfully using the product. The findings support visible, plain-language disclosure before consumers act, while also underscoring the importance of human review and cautioning that the survey was hypothetical, nonrandom and focused on ads created entirely with AI.

A disclosure shown before a purchase gives consumers information while they are deciding whether to trust the message. A disclosure that appears later serves a different purpose. It explains the production method after the consumer has already acted, when the information can feel like evidence that the brand withheld something relevant.

That timing issue extends beyond AI. Consumers often judge a business by how it handles information that could have affected a decision. A late explanation may be accurate and still arrive after the moment when it could have helped the customer make an informed choice.

The survey’s later-discovery scenario made that concern especially direct. Respondents were told that the advertisement had not disclosed its AI use when they saw it. That wording likely influenced the 70 percent who felt misled, and it means the study cannot isolate timing from the sense of nondisclosure. Even with that limitation, the practical lesson remains useful for advertisers: a disclosure placed after exposure or purchase carries a different trust signal from one presented before the ad does its persuasive work.

A small label buried in a terms page may also create a timing problem. If an ordinary consumer cannot see or understand the notice before responding to the ad, the brand may have technically provided information while failing to make it meaningful. Placement, wording and visibility all affect whether disclosure functions as consumer information or as a post hoc explanation.

The results support early disclosure while showing that disclosure alone has limits. Among people who saw the AI notice upfront, 57 percent still felt misled at least somewhat, and 62 percent felt manipulated at least somewhat.

Several parts of the scenario could explain those reactions. The ad persuaded respondents to make an unplanned purchase, which may have made the advertising feel aggressive. The scenario also said that AI created the entire advertisement, including both words and images. Consumers may react differently when AI assists with a minor task, such as resizing an image, than when it produces the complete creative work.

The survey did not ask respondents why they felt misled or manipulated. The results cannot identify whether those feelings came from the AI disclosure, the unexpected purchase, the ad’s persuasive approach or a combination of those factors.

That uncertainty gives brands a useful reason to look beyond a disclosure checklist. A clear notice can improve the information available to consumers, while the claims, imagery, targeting and emotional pressure in the ad still shape how the campaign feels. Consumers may accept AI production and remain uncomfortable with the way the ad uses it to persuade them.

The most useful lesson is not simply to add an AI label; it is to treat disclosure as part of the consumer decision itself. A notice that is visible, understandable and presented before the click or purchase gives people information when it can still influence their judgment. Brands should pair that disclosure with meaningful human review, because consumers want to know both how an ad was made and who is accountable for its claims, imagery and targeting.

The survey found broad support for two separate forms of transparency. Across both groups, 90.1 percent of respondents said it was very or somewhat important for brands to tell consumers when an advertisement had been created entirely using AI. Another 88.3 percent said it was very or somewhat important to know that a person had reviewed and approved the advertisement before publication.

Support for human review was almost identical across the two scenarios:

That consistency suggests that people view disclosure and review as related responsibilities with different purposes. Disclosure answers a question about how the content was made. Human review answers a question about who remains responsible for its claims and presentation.

A review process needs more substance than a final click in a publishing system. The responsible reviewer should be able to check whether the ad contains unsupported claims, misleading images, invented testimonials, inappropriate targeting or language that creates a false impression. The reviewer also needs enough authority to stop or revise the ad.

The European Commission’s explanation of transparency obligations under Article 50 of the EU AI Act provides one example of how responsibility can be framed in an organizational setting. Employees working under the authority and control of a legal person, such as advertising staff or content creators, are not treated as separate deployers simply because they operate the AI system. For brands, the broader point is clear: an organization using AI needs defined responsibility around that use.

The survey offers a useful directional signal, while its design limits the strength of any broad claim about all consumers.

ContentEngine used two separate online samples of 223 adults each. Respondents came from SurveyMonkey Audience panels, opted into online research and were not randomly selected from the U.S. adult population. The samples were not randomly assigned to the two scenarios, and the results were not weighted. A conventional margin of sampling error does not apply.

The samples also differed in the wording used to describe the experience. The later-discovery group was explicitly told that the ad had failed to disclose the AI use. That detail could have increased negative reactions independently of the timing itself. Similar measured demographics across the samples reduce one concern, but they cannot account for differences in unmeasured attitudes or experiences.

The scenario was hypothetical. Respondents did not see an actual advertisement, spend money or use a product. Stated trust and repeat purchase intent can differ from behavior in a live campaign. The findings also focus on advertisements whose words and images were created entirely with AI. They may not apply to a campaign where AI helped with a small production task or supported a human copywriter.

These limits call for careful language. The survey found an association between later discovery and weaker trust-related responses. It did not prove that delayed disclosure caused every difference. The association is still commercially important because advertisers often make decisions under uncertainty, and trust can erode before a company has perfect evidence about the size of the effect.

Brands can apply the findings through a few concrete operating rules.

The IAB’s AI Transparency and Disclosure Framework V2 takes a risk-based, materiality-driven approach to deciding when and how AI involvement should be disclosed. That type of framework can help companies avoid treating every use of software as identical. It also gives teams a way to connect disclosure decisions to the consumer’s likely understanding of the content.

Other research points to a growing visibility problem. A 2026 Bentley University-Gallup survey found that 79 percent of Americans had recently seen advertisements that appeared to be made using AI. A Canva study reported that 70 percent of consumers believed it would eventually become impossible to identify AI-generated ads without disclosure, while 56 percent expected that shift within two to five years.

As AI-made content becomes harder to identify by sight, consumers will have fewer reliable ways to judge how an ad was produced. Clear disclosures can fill part of that gap, but people will also look for signs that a real organization stands behind the output. Human review, accessible explanations and consistent accountability can make those signals more credible.

The ContentEngine survey puts the cost of delayed disclosure in practical terms. A campaign can generate a successful first purchase and a satisfied customer while weakening the customer’s willingness to trust or return to the brand. The safest point for a transparency decision is before the ad persuades someone to act.

For advertisers, early disclosure is a modest change in the creative and media process. Rebuilding trust after consumers discover hidden AI involvement is likely to require much more work, especially when the brand cannot explain who approved the message or why the disclosure came late.