Strong Brand Trust Isn’t Automatic With AI Ads: 36% Would Trust the Advertiser ‘A Great Deal’ in Consumer Research

A ContentEngine survey suggests that disclosure timing can shape how consumers judge an advertiser. Even after a satisfactory purchase, people who learned about an ad’s AI involvement later showed less trust and weaker interest in buying again.

The result centers on a simple scenario: an online advertisement persuades someone to make an unplanned purchase, and the buyer is satisfied with the product. In one version, the ad says before the purchase that its words and images were created entirely with AI. In the other, the buyer learns about the AI later and also learns that the ad did not disclose it at the time.

Among the people who saw the disclosure upfront, 36% said they would trust the advertiser a great deal. When the responses for a great deal and a fair amount are combined, trust reached 76.7%. That share fell to 65.0% among people who learned about the AI afterward, a gap of 11.7 percentage points.

The survey came from two online samples of 223 U.S. adults each, for 446 respondents overall. The findings describe the people who completed the surveys and offer a useful signal for brands testing AI-assisted creative. They do not provide a national estimate of consumer opinion.

A disclosure shown before a purchase gives consumers information while they are deciding whether to act. A disclosure encountered afterward arrives as an explanation of an earlier decision. Those situations carry different emotional and practical weight, even when the product performs well.

The later-discovery scenario made that difference especially clear by stating that the advertisement had failed to disclose its AI use. That language likely contributed to the increase in people who felt misled. It also reflects a real concern for brands: consumers may judge the omission itself, not just the technology behind the ad.

The report therefore treats the findings as an association between disclosure timing and consumer reactions. The wording did not isolate timing as a single causal factor. Respondents in the later group received a direct statement about nondisclosure, and the two groups were surveyed separately.

Still, the pattern has practical value. A brand deciding when to disclose AI involvement can expect the timing and wording of that disclosure to influence how consumers interpret the relationship. An early notice can be considered part of the buying experience. A later notice can be interpreted as a correction.

A ContentEngine survey of 446 U.S. adults found that consumers who were told an online ad was entirely AI-generated before purchasing reported higher trust, stronger repeat-purchase intent and fewer feelings of being misled than those who learned about the AI afterward. Trust reached 76.7% in the upfront-disclosure group versus 65.0% in the later-discovery group, while repeat-purchase intent differed by 12.6 percentage points. The findings suggest that meaningful AI disclosure works best before a consumer acts, but the survey’s separate nonprobability samples, hypothetical scenario and wording differences mean the results should guide campaign design rather than serve as a national estimate or definitive causal test.

A disclosure shown before a purchase gives consumers information while they are deciding whether to act. A disclosure encountered afterward arrives as an explanation of an earlier decision. Those situations carry different emotional and practical weight, even when the product performs well.

The later-discovery scenario made that difference especially clear by stating that the advertisement had failed to disclose its AI use. That language likely contributed to the increase in people who felt misled. It also reflects a real concern for brands: consumers may judge the omission itself, not just the technology behind the ad.

The report therefore treats the findings as an association between disclosure timing and consumer reactions. The wording did not isolate timing as a single causal factor. Respondents in the later group received a direct statement about nondisclosure, and the two groups were surveyed separately.

Still, the pattern has practical value. A brand deciding when to disclose AI involvement can expect the timing and wording of that disclosure to influence how consumers interpret the relationship. An early notice can be considered part of the buying experience. A later notice can be interpreted as a correction.

Every respondent was told that the purchase ended well. The product worked well enough for the buyer to feel satisfied, which gave the advertiser a favorable product outcome in both versions of the scenario.

That favorable outcome did not produce identical reactions. People who discovered the AI later still reported lower trust and weaker repeat purchase intent. The result suggests that product satisfaction and process transparency occupy separate places in the consumer’s judgment.

A customer can like a product and still question how the sale was made. That question may affect the next purchase, a recommendation to a friend or the customer’s willingness to believe future claims. The survey measured stated reactions, so actual marketplace behavior could differ, but the split is important for teams that evaluate campaigns mainly through immediate conversion.

The creative condition was also strong: the entire advertisement’s words and images were described as AI-generated. Results for an ad that uses AI to resize an image, create versions of a human-written message or assist with research could look different. Brands should apply the findings most directly to campaigns with substantial AI involvement.

The most useful lesson is not that disclosure guarantees trust; it is that disclosure timing becomes part of the brand relationship. Telling consumers about substantial AI involvement before they decide gives them context, while revealing it afterward can make the information feel like a correction—even when the product performs well. Advertisers should pair an accurate, visible notice with real human accountability, but they should also test the claim, targeting and tone of the ad because transparency alone did not eliminate feelings of manipulation or being misled.

Across both samples, 402 of 446 respondents, or 90.1%, said it was very or somewhat important for brands to tell consumers when an advertisement had been created entirely using AI.

That support was slightly higher in the upfront group, at 92.8%, than in the later-discovery group, at 87.4%. The difference was not statistically significant after the survey’s adjustment for multiple comparisons. The broad level of support is the more durable signal: consumers in both scenarios wanted information about how the ad was produced.

The survey also showed that disclosure alone may leave questions unresolved. In the upfront group, 57% said they felt misled at least somewhat, and 62.3% said they felt manipulated at least somewhat. The later group reported 70.0% feeling misled and 70.4% feeling manipulated.

Several features of the scenario may explain those reactions. The purchase was unplanned, the ad was persuasive enough to change behavior, and the entire creative was AI-generated. A respondent could object to the persuasive pressure, the surprise purchase, the use of AI or the combination of those elements.

The survey did not ask respondents to explain their answers. That limits the conclusions about why people felt manipulated or misled. It does establish that a label by itself may not settle every concern. Brands still need to examine the claim, targeting and tone of an advertisement.

Human oversight received nearly the same level of support as AI disclosure. Overall, 88.3% of respondents said it was very or somewhat important to know that a person had reviewed and approved an AI-generated advertisement before publication.

The result was almost identical across the two groups: 87.9% among people who saw the upfront disclosure and 88.8% among people who learned about AI afterward. That one-point difference was the smallest measured gap in the survey.

Disclosure and review answer different consumer questions. A disclosure tells people how an advertisement was made. Review identifies who remains accountable for its claims, images and implications. A label can describe production while a documented review process supports responsibility.

For an ad team, meaningful review could include checks such as:

A person clicking an approval box without examining the output provides little assurance. The survey does not test specific review procedures, but its results give brands a reason to define the role clearly.

The most direct operational lesson is to decide disclosure before production begins. Teams should identify whether AI will generate copy, images, audio, video or significant variations, then determine how that use will be explained to the audience.

The notice should appear early enough to inform the relevant choice. For a social ad, that may mean placing the disclosure in the ad itself instead of burying it on a landing page. For a product page, it may mean keeping the information near the creative or claim that was generated. A post-purchase explanation cannot perform the same informational role as a pre-purchase notice.

The language also needs to match the actual process. A statement that says an ad was created entirely with AI would be inaccurate if a human wrote the copy and AI produced image variations. A vague phrase such as AI enhanced may leave consumers unable to tell what the technology did.

A practical review sequence could look like this:

This process adds work before launch. It also gives brands a way to investigate complaints and update disclosures when the production method changes.

The ContentEngine research used two separate nonprobability samples recruited through SurveyMonkey Audience. Respondents opted into online panels, and the results were not weighted. A conventional margin of sampling error was not reported because the sample was not a probability sample.

Respondents also were not randomly assigned to the two scenarios. The measured demographic profiles did not differ significantly, but demographic similarity cannot rule out differences in attitudes, technology experience or sensitivity to advertising. Those unmeasured differences could contribute to the results.

The later-discovery scenario included an explicit statement that the ad had not disclosed AI use. That wording directly points toward deception and could increase reports of feeling misled. The survey also used a hypothetical advertisement. Respondents did not view a real ad, spend money or use a real product.

The results cover an ad made entirely with AI. They may not transfer to smaller uses of the technology. The survey did not measure race, ethnicity or education, and it did not ask people to explain their responses. These limits make the study better suited to guiding questions for campaign design than to predicting every consumer’s behavior.

The clearest reading of the findings is practical: brands should disclose substantial AI involvement before the consumer makes the relevant decision, and they should identify a person who is responsible for reviewing the final work.

That standard does not guarantee a positive reaction. The upfront group still reported meaningful levels of feeling misled and manipulated. Transparency gives consumers information, while responsible creative decisions determine whether the ad feels fair, accurate and proportionate.

For brand teams, the 36% great-deal trust figure is a useful warning against treating AI disclosure as a minor production footnote. Trust depends on the strength of the relationship, and the survey suggests that timing can weaken that relationship even after a customer likes the product.