AIRU

AI9 min read

Do Customers Trust Brands With AI-Written Blogs? 2026 Survey Data

Surveys in 2026 say more customers trust a brand less for AI-generated content than trust it more, and a belief that AI wrote a page costs trust even when a person wrote it. The size of the penalty depends on who ran the poll.

Hands typing on a laptop keyboard seen from above
Readers react to what they think the writer was, so the keyboard matters less than the byline.

Do customers trust brands whose blog content is written by AI?

On balance they trust them less. Meltwater and YouGov polled 9,869 adults in seven markets in early 2026 and found 32% would trust a brand less for AI-generated content against 15% who would trust it moremarket. Gartner asked 1,006 US consumers in May and June 2026, and 57% said AI-generated brand content made them less trusting of brand messagingmarket.

The UK reads worse. Optimizely found 62% of UK consumers trust marketing content less since brands started using AImarket. Wording matters here: "less trusting" is an attitude, and none of these polls measured what people did afterward. Before any of that, check whether you own the blog posts your agency delivers.

Does it matter if the AI label is true?

Not much. Raptive showed 3,000 US adults articles and adsmarket. It told one group the text was human-written and another that it was AI-made, and the groups got identical content. Readers who believed AI wrote it rated the content 48% less trustworthymarket.

A second poll points the same direction from the brand side. WordPress VIP reports that 60% of US consumers call brands using the word "AI" in their messaging a turnoffmarket. Both numbers measure a reaction to the label, so a company that hides AI use and gets found out carries the same penalty and the added problem of a trust story that has already failed. Trust signals like these now shape any blog strategy for the AI era.

Does AI content change what people buy?

Stated buying intent drops, in two separate polls. Products advertised next to content suspected of being AI saw purchase consideration fall 14%market. Clutch asked 601 US consumers about AI in social content, and 53% said they were less likely to buy from brands that use itmarket.

Neither poll tracked real purchases. A drop in stated intent is the upper bound of the effect. Real behaviour is usually softer, though no source in this set shows by how much. For readers deciding whether blog posts feed pipeline, the more useful view sits in what makes ChatGPT cite a company, because a reader who never sees the page has no opinion about it.

Can readers actually tell AI-written content?

More than half say they can. 57% of UK consumers say they can usually tell when content was generated by AImarket, and 58% of consumers across seven markets say the samemarket.

Confidence is a claim, not a test result. The Raptive design above shows why it matters even when it is wrong: people marked down text they merely believed was machine-made. Perceived AI, true or false, is what the reader acts on.

Is the problem AI itself, or too much weak content?

Volume draws the loudest complaint. 66% of UK consumers think brands prioritise volume over qualitymarket, and 65% of US consumers say brands produce too much AI-generated contentmarket. Quality gets named as well: 49% of US consumers say generative AI has made content quality worsemarket, and 59% of UK consumers say marketing content quality has declined in recent yearsmarket.

Attitudes toward AI itself are split. Meltwater and YouGov report 51% skeptical and 39% excitedmarket, and Clutch found 52% view AI positively overallmarket. People tolerate the tool and punish content that reads as filler. A site that publishes few, sourced pages is in a different position from one that buys guest posts at scale.

How do the six surveys compare?

They agree on direction and differ on size. The table sets out who asked, who paid and what each found.

SurveySample and dateHeadline findingWho ran it
Meltwater and YouGov9,869 adults, seven markets, Feb-Mar 202632% trust less, 15% trust moremarketMedia-monitoring vendor with a polling firm
Gartner1,006 US consumers, May-June 202657% less trusting of brand messagingmarketResearch and advisory firm
Optimizely1,000 UK consumers, Sept 202662% trust marketing lessmarketMarketing software vendor
Raptive3,000 US adults, 202548% lower trust when AI is suspectedmarketPublisher ad network
Clutch601 US consumers, Aug 202653% less likely to buymarketAgency directory
WordPress VIP1,200 US adults, April 202660% call "AI" in messaging a turnoffmarketPublishing platform vendor

Most of these polls were run by companies that sell something related to content, which is a reason to read each figure as a direction. The gap between the Meltwater and Optimizely headline numbers mostly reflects different questions and different countriesestimate, so averaging them gives nothing meaningful.

What should a company blog do with this data?

Put a named person on the page and say what the person did. 86% of consumers want AI-generated content disclosedmarket, and nearly 90% of US respondents in the Clutch poll say brands and creators should disclose AI-generated social contentmarket. Disclosure also takes the sting out of a later discovery.

Then cut volume before you cut AI. Readers complain about filler, so every page needs a fact the reader can check, which is the same property that gets a page cited. If a page already misstates your business, fix that before publishing more, starting with how to correct wrong ChatGPT answers.

To see whether any of this reaches buyers, a paid AI visibility tracker can show whether assistants mention you at all.

Trophies and a framed portrait on a wooden shelf
Credit goes to a named person, and readers look for that name.
Several sports trophies displayed on indoor shelves
Trust is built page by page and lost faster than it is earned.

Frequently asked questions

Do readers care if a blog post is AI-written?

More say it lowers trust than raises it. In the Meltwater and YouGov poll, 32% trust a brand less and 15% trust it moremarket. Gartner's US poll found 57% less trusting of brand messagingmarket. Both measure attitudes, not behaviour.

Should I disclose that AI helped write a blog post?

Yes, if you want to match what readers say they expect. 86% of consumers say AI-generated content should be disclosedmarket, and Clutch respondents show nearly 90% on social contentmarket. Name the person who edited and checked the facts.

Can people tell when a company used AI to write a post?

Most believe they can, and belief is enough to change their view. 57% of UK consumers say they can usually tellmarket, and 58% across seven markets agreemarket. Raptive found content was rated lower whenever readers merely believed AI wrote it.

Does AI content reduce sales?

Stated intent falls, and measured sales are unknown. Products next to suspected AI content had 14% lower purchase considerationmarket, and 53% of Clutch respondents are less likely to buy from brands using AI in social contentmarket. Neither poll counted purchases.

Is the problem AI or low-quality content?

The data points to quality and volume. 66% of UK consumers say brands prioritise volume over qualitymarket, and 65% of US consumers say brands publish too much AI-generated contentmarket. A few well-sourced pages avoid the complaint better than many thin ones.

Why do these surveys disagree?

They ask different questions of different groups. A question about "less trust" gets 32% in one pollmarket and 62% in anothermarket, and the sponsors sell content tools, ad space or agency listings. Read the direction across polls and discount any single figure.

Sources

  1. Meltwater and YouGov, Trust in the Age of Generative AI, 9,869 adults in seven markets, fieldwork 26 Feb-5 Mar 2026: 32% trust less vs 15% more, 86% want disclosure, 58% believe they can identify AI content, 51% skeptical vs 39% excited. https://www.meltwater.com/en/press-releases/consumer-perception-of-generative-ai (April 2026)
  2. Clutch, consumer survey of 601 US consumers on AI in social media content: nearly 90% want disclosure, 53% less likely to buy, 52% view AI positively. https://letsdatascience.com/news/clutch-reports-consumer-distrust-of-ai-social-content-ea7c1666 (August 2026)
  3. Gartner, survey of 1,006 US consumers: 57% less trusting, 65% too much AI content, 49% say quality got worse. https://www.gartner.com/en/newsroom/press-releases/2026-06-09-gartner-survey-finds-49-of-u-s-consumers-say-genai-has-made-content-quality-worse (June 2026)
  4. Gartner, newsroom release on consumers and AI. https://www.gartner.com/en/newsroom/press-releases/2026-09-22-gartner-marketing-survuey-finds-35-percent-of-consumers-rely-on-influencers-less-due-to-ai (2026)
  5. Optimizely, UK survey of 1,000 consumers and 100 marketers: 62% trust less, 66% volume over quality, 57% can tell, 59% quality declined. https://www.optimizely.com/company/press/ai-trustworthiness-uk/ (September 2026)
  6. IT Brief UK, report on the Optimizely survey. https://itbrief.co.uk/story/uk-consumers-trust-ai-marketing-less-optimizely-finds (September 2026)
  7. Raptive, "The AI stink is real and it's costing brands," 3,000 US adults: 48% less trustworthy when AI is suspected. https://raptive.com/blog/the-ai-stink-is-real-and-its-costing-brands/ (July 2025)
  8. ppc.land, report on the Raptive study: 14% lower purchase consideration. https://ppc.land/raptive-study-shows-ai-content-cuts-reader-trust-by-half/ (July 2025)
  9. WordPress VIP survey, 1,200 US adults and 800 enterprise decision-makers: 60% call "AI" in messaging a turnoff, via ContentGrip. https://www.contentgrip.com/ai-brand-messaging-trust-survey/ (April 2026)

Check your site in a minute

Paste your address: we'll show your score out of 100, whether an AI assistant names you, and which searches your site has no page for. Free.