A global consumer study published today by RTB House reports that leading AI assistants now carry higher shopping trust scores than TikTok, Instagram, Facebook, newspapers and influencers, while 42% of United States respondents say those same tools stretch the time they need to settle on a purchase. The company surveyed 1,840 people in the United States, the United Kingdom, France and Japan during June and July 2026.

The study, titled "Who's Buying? Consumer Trust in the Age of Agentic AI," was released today by RTB House, the Warsaw-founded performance marketing company. Fieldwork ran in June and July 2026 in partnership with Cint, the survey sampling platform. The sample compares United States respondents against a combined non-United States group drawn from the United Kingdom, France and Japan. All percentages in the published report are rounded to whole points, trust figures represent respondents selecting "mostly" or "very much," and agreement figures combine "somewhat agree" with "strongly agree."

Two findings sit in tension with each other. Consumers say they trust AI tools more than the social platforms that carried a decade of e-commerce growth. They also say those tools are making them slower to decide.

A new trust hierarchy, with an uneven distribution

Trust scores in the report favour two products above all others. Google AI Overviews records 43% trust among United States respondents, with ChatGPT level at 43%. Outside the United States the order reverses: ChatGPT leads at 41%, ahead of Google AI Overviews at 35%.

The gap to the next tier is wide. Claude registers 23% in the United States and 20% among non-United States respondents. Grok sits at 21% and 14% respectively. On the report's own framing, that spread separates the assistants shaping consideration sets from those that do not.

RTB House measured how far leading AI tools sit above established media channels in percentage points. Among United States respondents, the AI advantage runs to 20 points over influencers, 16 over newspapers, 15 over major media, 11 over both Instagram and Facebook, and 8 over TikTok. The non-United States figures show a larger divide almost everywhere: 25 points over influencers, 22 over Facebook, 18 over TikTok, 17 over Instagram, 16 over major media and 12 over newspapers.

One channel still outranks the machines. According to RTB House, 59% of United States consumers and 56% of non-United States consumers trust friends and family, against 44% for the leading AI tool. The report does not reconcile that 44% figure with the 43% recorded in its own trust chart, a one point discrepancy left unexplained.

Belief in AI neutrality remains a minority position. Across the United States sample, 42% agree that AI provides trusted, unbiased information, falling to 37% outside the United States. Millennials in the United States are the most likely group to agree, at 48%.

That reading is more measured than the trust rankings alone suggest, and it aligns with adjacent research. A Yelp study published in April 2026 found only 15% of users trust AI search results, with 63% double-checking those results against news sites and review platforms. Reddit's Path to Purchase survey, covered in June 2026, found half of United States shoppers verifying AI product recommendations on the platform before buying. Relative trust and absolute trust are not the same measurement, and the RTB House data captures the former.

The consideration window is getting wider

The central claim of the report concerns time. Respondents were asked to react to a single statement: "AI tools increase the time it takes me to make a final decision because they provide more information, new choices, and a larger set of products or brands to consider."

In the United States, 29% somewhat agree and 13% strongly agree, producing the 42% headline figure. Among non-United States respondents, 26% somewhat agree and 6% strongly agree. That sums to 32% in the report's chart, though the accompanying press materials describe the non-United States figure as 33%.

Generational patterns in the United States show the effect concentrated among the youngest cohort. RTB House reports 48% agreement among Gen Z, followed by 42% of baby boomers, 40% of Gen X and 40% of Millennials. Outside the United States the distribution is flatter: 35% of Millennials, 34% of baby boomers, 33% of Gen Z and 32% of Gen X.

Those narrative figures do not match the stacked bar chart printed directly beneath them. The chart labelled United States shows totals of 33% for Gen Z, 35% for Millennials, 32% for Gen X and 34% for boomers, while the panel labelled non-United States shows 48%, 40%, 40% and 42%. The two panels appear to carry each other's labels. The narrative text is internally consistent with the report's own headline numbers, so the chart labelling is the more likely error, but RTB House does not flag the inconsistency.

Site visit data provides the mechanical context. According to the report, United States shoppers most commonly visit a site four to six times before completing a purchase. The report frames AI as widening that window rather than closing it: AI is not simply shortening the path to purchase but expanding the consideration phase by bringing more alternatives to a shopper's attention.

A separate line under the same section states that 42% of Gen Z shop fashion exclusively on mobile, nearly three times the rate of boomers. No supporting chart, question wording or cross-tabulation for that claim appears anywhere else in the 14-page document.

Discovery skews older in the United States

The brand discovery numbers cut against the assumption that AI shopping behaviour belongs to younger cohorts. Across the United States sample, 59% agree that AI is good at surfacing brands they did not previously know, against 52% outside the United States.

Broken down by generation, agreement in the United States rises with age. Baby boomers record 62% and Gen X 61%, both ahead of Gen Z at 52%. The non-United States pattern inverts: Gen Z leads at 55%, with Gen X at 49% and boomers at 50%. The widest United States to non-United States gaps therefore appear among older consumers, while Gen Z shows near parity across geographies.

Usage backs the discovery claim. RTB House reports that 63% of United States respondents have used AI to turn a shopping need into at least a general list of products.

These figures extend a consideration-stage picture RTB House has published before. A study the company released in March 2026 found that only 40% of new customers arrive at e-commerce sites with a specific item in mind, and that 50% of Gen Z shoppers require two or more days of consideration after a product enters the cart. The August data adds an upstream layer: the assistant now populates the shortlist before the site visit happens.

Delegation, and the price of removing the safety net

The report's agentic commerce section tested a specific financial threshold. Respondents were asked whether they would be comfortable giving an AI tool their own money, up to $250, to purchase an item based on stated requirements, assuming a seven-day return window.

With that return window in place, 42% of United States Millennials say yes. Gen Z follows at 35%, Gen X at 29% and baby boomers at 22%. The non-United States figures run lower across every cohort: Millennials at 27%, Gen Z at 26%, Gen X at 22% and baby boomers at 15%. The pitch materials accompanying the release convert the $250 threshold to £186.

Remove the return guarantee and the numbers fall unevenly. United States Millennials drop to 34%, Gen Z to 32%, Gen X to 24% and boomers to 11%, a decline of roughly eleven points that the report identifies as the largest of any group. Outside the United States, Millennials fall to 20%, Gen Z to 25%, Gen X to 15% and boomers to 8%. Return protection, in other words, functions as the load-bearing element for older shoppers and a secondary consideration for younger ones.

Asked which conditions would make them more comfortable letting an AI agent purchase on their behalf, respondents ranked human approval before checkout first at 35% globally. The ability to cancel immediately after purchase follows at 31%, purchasing only from trusted retailers at 29%, a strict price limit at 29%, and free or simple returns at 28%.

Baby boomers request the most oversight. According to RTB House, 44% of boomers want human approval before checkout, 35% want purchases restricted to trusted retailers, and 33% require a strict price limit. Preference for trusted retailers reaches 32% among Millennials and 31% among Gen X.

Here too the document contains an internal inconsistency. The report puts free or simple returns at 28% overall, then describes generational results for the same item ranging from approximately 30% among Millennials to 35% among baby boomers. A range that sits entirely above the stated overall figure cannot be produced by the same question without additional explanation, and none is given.

What RTB House says the data means

Jaysen Gillespie, VP of Product Marketing and Analytics at RTB House, framed the findings as the end of a commercial era.

"For more than a decade, e-commerce growth was driven by buying attention on social feeds and optimising for instant checkouts," Gillespie said. "This research proves that era is coming to an end. Consumers are turning to AI as an analytical powerhouse, using it to evaluate and discover brands, conduct price comparisons, and conduct deeper research before they buy. As AI expands the consideration window, competition is fiercer than ever."

On where budget should follow that behaviour, Gillespie added: "Modern marketers will find that while discovery happens in AI engines, the rubber meets the road at on-site and in-app monetisation. The longer research window gives marketers an opportunity to more deeply engage with timely and relevant offers across the entire shopper set. Our advice would be to take advantage of that extended window."

The commercial interest behind that framing is not hidden. RTB House sells retargeting and on-site engagement technology, including rtb.com, the self-service platform it opened to smaller e-commerce brands in March 2026. A longer consideration window is, mechanically, more retargeting inventory. The report's own guidance to marketers states that sustained engagement and retargeting become more important as AI expands the research window.

Why the marketing community is watching this data

The findings land in a market where the infrastructure for agentic purchasing has been built faster than the consumer appetite for using it.

Google launched agentic checkout alongside conversational shopping tools in November 2025, then introduced shopping ads inside AI Mode in February 2026 as that surface passed 75 million daily active users. Microsoft launched Copilot Checkout on 8 January 2026. Amazon, Meta, Microsoft, Salesforce and Stripe joined the Universal Commerce Protocol Tech Council on 24 April 2026, consolidating governance of the standard that defines how agents interact with merchants.

Consumer readiness has consistently lagged that build-out. Independent analyst Andrew Lipsman assigned a 25% probability to shoppers accepting products selected without direct inspection, one of four conditions he identified as necessary for agentic transactions to work. A Shift Browser survey published on 3 March 2026 found 81% of United States respondents concerned about AI accessing personal data, with 44% worried about AI taking unauthorised actions, and referenced a German industry framework finding only 25% of consumers there willing to delegate tasks to agents. Usercentrics research covering 11,000 consumers across seven markets, published in June 2026, quantified the commercial cost of that gap at 240,000 purchase-affecting decisions per million customers.

Against that backdrop, the RTB House delegation figures read as neither adoption nor rejection. A 42% comfort level among United States Millennials for a $250 supervised transaction is materially higher than the 25% delegation willingness recorded in Germany, and materially lower than the trust scores AI tools earn as advisors. The distance between the two is the guardrail.

The trust ranking carries a second implication for media planning. If AI assistants outrank TikTok, Instagram and Facebook as shopping advisors while sending a fraction of the referral traffic those platforms deliver, the influence is arriving in a channel that resists conventional measurement. Ahrefs research published on 4 February 2026 found ChatGPT sends roughly 190 times less traffic to websites than Google despite handling about 12% of its query volume. Google Analytics only added a dedicated AI assistant channel grouping in May 2026.

Where AI referrals do arrive, quality has been the argument. Criteo reported in May 2026 that conversion rates from ChatGPT-referred traffic approached twice those of traditional search in consumer electronics, lifestyle and home categories. In its most recent quarterly disclosure, the company reported that roughly 80% of paid traffic arriving through the ChatGPT integration was new to the brand. Those figures come from a vendor's own client reporting rather than independent audit, but they point in the same direction as the RTB House discovery data: shoppers reaching a site through an AI assistant tend to arrive further along, and often unfamiliar with the brand they are considering.

The counterweight sits on the publisher side of the same equation. A randomised field experiment with 1,065 Chrome users found that AI Overviews cut outbound organic clicks by 39.8%, while Ahrefs research from February 2026 correlated the feature with a 58% reduction in click-through rates for top-ranking pages. A trusted advisory surface that answers in-stream produces exactly the pattern the RTB House data describes from the consumer side: more brands considered, more information absorbed, fewer clicks distributed, and a decision that takes longer to make.

Methodology and what the study does not show

The sample is 1,840 respondents split between the United States and a combined group covering the United Kingdom, France and Japan. RTB House does not publish per-country sample sizes, so the non-United States figures cannot be separated into British, French and Japanese components. Three markets with materially different e-commerce structures are reported as a single unit throughout.

The study measures stated attitudes, not observed behaviour. Willingness to hand $250 to an agent under a hypothetical return policy is a survey response, not a transaction. Comfort levels of this kind have historically overstated adoption in agentic commerce research, a pattern documented in coverage of consumer readiness across 2025 and 2026.

The press materials issued alongside the report also state that 46% of United States Millennials have already purchased from a brand they discovered through AI this year. That figure appears in the release headline bullets but not in the published 14-page report, which contains no supporting chart, question wording or cross-tabulation for it. RTB House does not otherwise document purchase completion rates attributable to AI discovery.

Neither the report nor the release addresses which categories the delegation figures apply to, whether the $250 threshold was tested against alternative amounts, or how respondents were screened for prior AI usage. Those omissions limit what the delegation data can support beyond a directional reading.

Timeline

Summary

Who: RTB House, a performance marketing technology company founded in 2012 and operating more than 3,000 active campaigns across EMEA, APAC and the Americas, published the research. Jaysen Gillespie, VP of Product Marketing and Analytics at RTB House, provided the company's commentary. Fieldwork was conducted by Cint.

What: A consumer study titled "Who's Buying? Consumer Trust in the Age of Agentic AI," finding that leading AI tools now outrank social platforms, newspapers and influencers on shopping trust, that 42% of United States consumers and 32% of non-United States consumers say AI lengthens their final purchase decision, that 59% of United States consumers credit AI with surfacing unfamiliar brands, and that 42% of United States Millennials would delegate a $250 purchase to an AI agent when a seven-day return window applies.

When: Fieldwork ran across June and July 2026. The report was published on 10 August 2026.

Where: The survey covered 1,840 respondents in the United States and a combined non-United States group spanning the United Kingdom, France and Japan.

Why: Platform operators have built agentic checkout infrastructure faster than consumers have adopted it, and the data quantifies where trust currently sits: high enough for AI assistants to outrank social feeds as shopping advisors, conditional enough that human approval before checkout remains the most requested safeguard globally, and structured in a way that widens rather than compresses the consideration phase that media buyers plan against.