How AI Recommendations Are Changing Which Brands People Buy: What the Survey Shows (and What It Can’t Prove)
A headline making the rounds this week says AI tools drive 64% of people to buy from new brands. For anyone who sells something online, that sounds like a warning and an opportunity at once. If AI assistants are introducing shoppers to brands they have never heard of, then being mentioned in an AI answer matters.
But a single statistic can mean very different things depending on who was asked, what they were asked, and who paid for the survey. This guide walks through what the survey reports, how the 64% was calculated, what independent research says, and what the evidence supports for businesses and shoppers.
What the Survey Reports
The story ran on October 22, 2026, in a news piece syndicated from Talker News. According to it, the survey was commissioned by We Are Talker and conducted online by Talker Research between September 21 and 30, 2026. It asked 2,000 Americans who had used an AI tool or assistant for a personal, non-work purpose in the previous three months. The article says the methodology follows the AAPOR Transparency Initiative and links to the questionnaire.
The reported results include:
- 70% discovered a new brand through AI, and 91% of those then bought or used something from it. The article’s headline figure of 64% comes from combining the two.
- 66% say they remember AI-recommended brands well.
- 51% would consider an unfamiliar brand if AI strongly recommended it.
- 68% say AI led them to rule out a brand they were considering.
- 44% say AI changed a decision they had planned to make.
- 81% followed an AI recommendation at least in part.
- 73% rate AI recommendations as influential, compared with 78% for customer reviews and 76% for friends and family.
- 48% use AI several times a week or more for purchase decisions.
The article does not name which AI tools respondents used. It quotes We Are Talker’s chief executive saying it is no longer enough simply to be visible.
I looked for the original report on Talker Research’s site and could not find it. Talker Research does have an earlier study, published in June 2026, on AI search, trust, and brand discovery, based on 1,000 Americans who use Google’s AI Overviews. It reports that 64% say their trust in AI has increased over the past year. That is a different study and a different 64%, so be careful not to mix them up when you see the figure quoted.
How the 64% Is Built
The arithmetic is simple: 70% of respondents discovered a new brand through AI, and 91% of those went on to buy or use something from it. Multiply them and you get roughly 64% of the sample.
Three details change how you should read that:
- The base is AI users only. The sample was limited to people who had used an AI tool in the past three months. It is not 64% of Americans or 64% of shoppers.
- “Bought or used” is not “bought.” The 91% includes people who used a product, not only those who paid for one.
- It is self-reported. Respondents said AI introduced them to a brand. Nobody checked whether the AI actually did, or whether the shopper would have found the brand some other way.
None of this makes the number false. It makes it a statement about what a group of AI users say they did, not a measurement of how much AI changes buying behavior.
Reasons for Caution
The sponsor sells brand services. The survey was commissioned by a firm that, according to the article, sells brand-marketing services. The main quotes come from its chief executive. That does not invalidate the data, but it is a reason to read the framing with care. The article does not label itself as sponsored or as a press release.
The sample is not described as representative. The article does not say how respondents were recruited or whether the results were weighted to match the US population.
Question wording is not visible in the article. The way a question is asked can move an answer by many points. The questionnaire is linked, but I could not open the original.
The AI tools are unnamed. A recommendation from a general chatbot, a shopping assistant, and an AI summary in a search engine are different experiences.
“Influential” is a feeling. Seventy-three percent rating AI recommendations as influential tells us how people perceive AI, not how often it decides a purchase.
What Independent Research Says
Two other sources give useful context. Both have their own limits.
A survey that points the other way
A 2026 study from the PR firm Idea Grove, run by Pollfish with 1,000 US consumers and reported by Chain Store Age, found that only 2% would buy from an unfamiliar brand based solely on an AI recommendation. The other 98% would take additional steps first. After an AI suggestion, 45% said they would immediately search for the brand on Google and 18% would go straight to review sites. Seventy-eight percent said customer reviews increase trust, and 85% expressed some skepticism about AI recommendations. Nearly half did not realize companies try to influence AI recommendations. The fielding dates were not stated. Like the Talker survey, it comes from a firm with a commercial interest in the topic.
A large consultancy study
Boston Consulting Group, in a January 2026 report, drew on consumer survey waves of more than 9,000 respondents in each of two years across nine countries, plus video-based research with 303 US consumers who had recently used generative AI in a purchase journey. BCG reports that shopping-related use of generative AI rose 35% between February and November 2025. It says AI’s reach is still lower than search, brand sites, and social media, but it is often highly influential, and it can steer shoppers toward specific brands while they pursue a broader goal such as meal planning or home repair. More than 60% of respondents expressed high trust in generative AI results.
Putting the studies side by side
| Source | Who ran it | Sample | Headline | Caution |
|---|---|---|---|---|
| Talker (October 2026) | Talker Research for We Are Talker | 2,000 US AI users | 64% bought from a new brand AI introduced | Self-reported; sponsor sells brand services; original not found |
| Idea Grove (2026) | Idea Grove, run by Pollfish | 1,000 US consumers | Only 2% would buy from an unknown brand on AI advice alone | PR firm with a stake; dates not stated |
| BCG (January 2026) | Boston Consulting Group | 9,000+ per wave, 9 countries, plus 303 US shoppers | AI shopping use up 35%; influence high though reach is lower than search | Broader and more cautious about reach |
At first glance, 81% following an AI recommendation and 98% double-checking look contradictory. They are not necessarily. “Followed at least in part” is a much looser behavior than “bought an unfamiliar brand based only on AI.” People can follow a suggestion and still verify it first. The common thread is that AI is often the start of the journey, and search and reviews often finish it.
What This Means for Small Businesses
The most reliable takeaway is not “get mentioned by AI at any cost.” It is that discovery and verification now happen in sequence: a shopper may meet your brand in an AI answer, then search for you and read your reviews before buying. That suggests three practical steps.
- Be easy to find when someone searches your name. Check what appears for your brand name, and make sure your site and business listings are accurate.
- Collect honest reviews. Both the Talker article and the Idea Grove study point to reviews as a key trust factor.
- Do ordinary good SEO. Google’s guidance says the same SEO best practices apply to AI Overviews and AI Mode as to Search overall, and that no special optimization, new files, or special markup is required. It recommends allowing crawling, keeping important content in text, matching structured data to visible text, and keeping Merchant Center and Business Profile information current. Google also says meeting these requirements does not guarantee a page will be crawled, indexed, or shown.
You will see many agencies selling “AI visibility” services. Google’s own guidance suggests there is no secret formula for its AI features. Other AI assistants may behave differently, and their recommendation methods are less documented, so treat bold promises with caution. For tools that can help with the everyday work of content and marketing, see our roundup of AI tools for marketers that still work, and if you sell handmade goods, our guide to AI tools for Etsy sellers.
What This Means for Shoppers
If AI is introducing you to brands, a few habits help:
- Treat a recommendation as a lead, not a verdict. Search the brand and read independent reviews.
- Ask for reasons and sources. A good answer explains why a product fits and where the claim comes from.
- Remember that someone may be trying to influence results. Almost half of respondents in the Idea Grove study did not realize that companies try to shape AI answers.
- Compare more than one option. Ask the assistant for alternatives and what each does worse.
Common Mistakes to Avoid
Quoting the 64% without its base. It describes AI users who said they discovered a new brand and then used it, not all consumers.
Treating a sponsored survey as neutral. Check who paid and what they sell.
Confusing two different 64% figures. Talker’s June study found 64% saying trust in AI had grown. The October survey’s 64% is about buying from new brands.
Assuming AI recommendations replace reviews. Several sources show shoppers still verify.
Paying for “AI ranking” promises. Google says no special optimization is required for its AI features.
Ignoring the basics. Accurate listings, helpful content, and honest reviews still do the heavy lifting.
Frequently Asked Questions
Does AI really make people buy new brands? Some AI users say it introduced them to brands they then bought or used. The evidence is self-reported, and other research shows most shoppers verify before buying.
What does the 64% mean? It combines 70% who discovered a new brand through AI with 91% of those who then bought or used something from it, among 2,000 US AI users.
Who paid for the survey? According to the article, We Are Talker commissioned it and Talker Research conducted it.
Can I pay to be recommended by AI? Google says there are no extra requirements or special optimizations for AI Overviews or AI Mode. Be cautious about services that promise guaranteed placement.
Which AI tools were used? The article does not say.
Should I trust AI product recommendations? Use them as a starting point, then check reviews and independent sources.
The Bottom Line
AI assistants are becoming a place where shoppers meet brands, and surveys agree on that much. What they disagree on is how often an AI suggestion turns directly into a purchase. The 64% figure is a real survey result, but it comes from a sponsored study, a self-reported sample of AI users, and a calculation that includes people who only “used” a product. Independent research suggests that most people still search and read reviews before they buy. For businesses, the safest plan is the unglamorous one: be findable, be trustworthy, and keep your information accurate everywhere shoppers look.
Sources
- Wyoming News Now (from Talker News): Survey finds AI tools drive 64% to buy new brands
- Talker Research: AI search, trust and brand discovery study (June 2026)
- Chain Store Age: consumers rely on online reviews to vet AI product suggestions
- Boston Consulting Group: consumers trust AI to buy, better brands must adapt (January 2026)
- Google Search Central: AI features and your website











































