The AI Visibility Brief: June 15-21, 2026

THIS WEEK

• Getting cited does not mean a brand will be recommended
• New survey data: AI use is rising while trust is falling
• Google says there is no "format for AI" trick
• New free AI reports from Bing and Google

01

Getting cited does not mean a brand will be recommended

Lily Ray analyzed 100 "best [category] software" questions inside Google's AI Overviews, checked across three dates this spring. Of the 80 questions that returned an AI Overview, brands' own self-ranked "best of" lists were cited 323 times. But in 224 of those cases, about 69%, Google pulled its quote from the brand's page and then recommended a different company in the answer, sometimes one the brand itself had ranked further down its own list. The brands that did get recommended were the category leaders with broad third-party coverage and stronger link profiles. Forbes, Reddit and YouTube were among the most-cited sources.

We think this matters because a citation and a recommendation are not the same thing. A citation is a source link, usually listed as where the answer came from. A recommendation is when the AI names your brand in the answer itself as one to choose. A buyer is far more likely to act on the brand AI models recommend than on a link in their source list.

Businesses should act on this by earning the recommendation through independent coverage, reviews and mentions across the sites these tools pull from, which is what set the recommended brands apart in Ray's data. It is no surprise AI models will not take your word that you are the best when you are the one ranking yourself first on your own page.

02

People are using AI search more and trusting it less

That is the headline finding from a Fractl survey of 1,008 U.S. consumers and 150 marketers, published June 17. It is self-reported, so it captures what people say rather than what they actually do. In it, 70% report using AI search more than they did a year ago. Over the same year, the share who say AI search is more helpful than traditional search fell from 82% to 54%, and the share who say they would trust a brand less if it leans heavily on AI-written content rose from 20% to 39%. Before making a purchase, people said they check about 2.4 different platforms.

We think this matters because being recommended in an AI answer and having consumers trust that answer are two different things. Usage keeps climbing while trust slips, so being shown in an AI answer is only part of it. People also have to trust what that answer says about you.

Businesses should act on this by being present and consistent across Google, Reddit, YouTube and review sites, since buyers said they check about 2.4 different places before they buy. A brand that only looks good on its own website fails that cross-check.

03

Google says there is no "format your content for AI" trick

At Google Search Central Live in Milan on June 18, Google cleared up several AI-search questions. On the popular idea of breaking content into special chunks for AI (splitting your text into small, self-contained blocks so AI models can lift them out more easily), the takeaway was blunt: forcing paragraph chunking for AI is useless, and content should be organized for human readability. Google said what it rewards is specific, original content based on real first-hand experience, and that mass-produced, generic AI text with no original information is treated as spam. It also confirmed that for more complex questions it now runs several searches in parallel behind the scenes before answering.

We think this matters because it puts a popular formatting hack to rest and points back to substance. The signal Google keeps repeating is original, first-hand content, and restructuring what you already have is not a substitute for it.

Businesses should act on this by putting the effort into detailed, first-hand content drawn from real work, like write-ups of actual jobs done, and making sure it lives on pages AI can read, since that is the kind of specific, first-hand material Google says it rewards.

04

Bing and Google now offer free ways to see where you stand in AI answers

On June 16, Microsoft added an expanded AI report inside Bing Webmaster Tools with a new Citation Share metric that shows how often a site is cited in Bing's AI answers, and how that compares with other sites. On June 18, Google began rolling out a Search Console AI report that separates AI Overviews, AI Mode and Discover clicks and impressions, along with a setting to include or exclude a site from AI features.

We think this matters because until recently, site owners had almost no direct read on how they show up in AI answers. Now two of the biggest platforms are giving that data away for free. The catch is that each one only sees its own platform, so no single report shows how a brand is doing across AI as a whole.

Businesses should act on this by treating these as free starting points and checking Google's new include or exclude setting first, since flipping it the wrong way would pull a website out of AI features entirely. Neither report says anything about how a brand shows up in ChatGPT or Perplexity, so those still need checking on their own.

Thanks for reading this week’s newsletter.
Sincerely,
Brett