[The Engines]
[Example] Google AI Overviews expands: what changed and what to do
Google's biggest AI Overviews overhaul since launch rewired how sources appear in the SERP. Here is what changed, who it affects, and what to do about it.
The short answer
On May 6, 2026, Google rolled out five structural changes to AI Overviews: inline citations, hover previews, an Expert Advice block, subscription labels, and an Explore New Angles section. The feature now covers more than 200 countries and roughly 48% of US queries, up from 6.49% in January 2025. Brands that earn citations in AI Overviews capture approximately 120% more organic clicks per impression than uncited brands on the same query.
What exactly changed in Google AI Overviews?
On May 6, 2026, Google deployed the most significant structural overhaul to AI Overviews since the product launched. Five distinct surface changes landed at once, each shifting how the SERP presents sources, experts, and follow-up paths.
The first and most consequential change: citations moved inline. Previously, every source was grouped into a carousel at the bottom of the AI Overview box. Now each claim links directly to the source that supports it, sitting right beside the sentence or paragraph it backs. That is a structural shift in how Google attributes information inside the answer layer, and it has direct consequences for which brands earn visible credit and which disappear into a collapsed list no one scrolls to.
Second, hover previews now show the site name and page title on desktop before the user clicks. Source identity is no longer hidden behind a favicon. A user reading about project management tools can see whether the cited piece comes from a respected publication or an obscure thin site without leaving the page. Trust signals built into the brand now travel into the SERP surface itself.
Third, Google introduced an Expert Advice block that surfaces first-hand perspectives from Reddit, forums, and social media, with creator names and handles displayed. Community content, long treated as a secondary signal and excluded from most editorial SEO strategies, now has a named slot inside the AI Overview. This is a net-new citation surface that did not exist six months ago.
Fourth, a Subscribed label highlights links from publications the user already pays for. Paywalled journalism becomes more accessible inside the AI response rather than invisible behind a lock icon. For publishers competing on authority, this is a distribution layer that rewards existing subscriber relationships.
Fifth, an Explore New Angles section appears at the end of many AI responses, surfacing follow-up topic links: deeper subtopics, case studies, comparisons. Google is building a structured recommendation layer into the AI Overview itself, creating demand for specific, narrow content rather than broad authority pieces. Each of these changes points in the same direction: specificity and source identity matter more than they did before this update.
How much has the AI Overviews footprint grown?
The May 2026 feature update landed on top of an already rapid geographic and query-coverage expansion. In January 2025, AI Overviews appeared in approximately 6.49% of US Google searches. By mid-2026 that figure sits at roughly 48% of all US queries, according to data compiled by The Stacc. That is a nearly eight-fold increase in eighteen months.
Internationally, Google announced at I/O 2026 that AI Overviews now reach more than 2.5 billion users per month across 200 countries and territories and 40 or more languages. The May 2025 expansion added Arabic, Chinese, Malay, and Urdu; a February 2026 update added 53 additional languages in a single release, the largest single-language addition to date.
Query-type coverage has also broadened well beyond where the product started. In January 2025, 91.3% of AIO-triggering queries were informational. By October 2025 that share had dropped to 57.1%, with commercial queries growing to roughly 19% and navigational triggers rising from under 1% to over 10%. AI Overviews is no longer a product for open-ended research questions. It is increasingly present on decision-stage and brand-comparison queries, exactly the prompts where your category buyer is choosing between you and a competitor.
That trajectory means the competitive surface is growing faster than most brands realize. If you audited AI Overviews coverage in your category a year ago and concluded it was an edge case, the data says run the prompts again. The queries that matter to your pipeline are more likely to trigger an AI Overview today than at any point in the feature's history.
| SERP element | Before May 6, 2026 | After May 6, 2026 |
|---|---|---|
| Citation placement | Grouped in a collapsed carousel at the bottom of the AI Overview box | Inline, next to the specific sentence or claim each source supports |
| Source identity on desktop | Domain shown only after the user clicks through | Site name and page title visible on hover, before any click |
| Community content | Not surfaced inside the AI Overview box | Named Expert Advice block: Reddit, forums, social platforms, with creator handles shown |
| Paywalled publications | No distinction from free content in the source list | Subscribed label on links from publications the user already pays for |
| Follow-up discovery | No structured path beyond the AI box itself | Explore New Angles section surfaces subtopics, case studies, and comparisons at the end of the response |
Who does the May 2026 update actually affect?
The inline citation change affects anyone who currently earns a slot in the AI Overview box. Your citation now carries your brand name visibly, next to specific claims, before the user clicks. That is a brand-impression opportunity that did not exist in the previous format, where citations were buried in a collapsed carousel most users never opened.
The Expert Advice block creates a new citation surface specifically for brands with community presence. If your team participates substantively in forums, writes detailed Reddit threads, or publishes first-hand experience on social platforms, that content can now surface inside the AI Overview without owning a traditional editorial link in the main body. For B2B SaaS companies whose buyers live in communities like specialist subreddits or industry Slack groups, this is a pathway into the AI answer layer that bypasses traditional link-building entirely.
The Explore New Angles section creates structured demand for depth. Google is surfacing subtopics, case studies, and comparisons inside the response rather than leaving discovery to the user. Brands with a library of tightly scoped, specific content now have more points of entry into AI-surfaced follow-up paths, not just the primary answer. A single broad pillar page earns one citation at best. Fifty articles each answering one narrow question create fifty potential citation surfaces.
The brands most exposed negatively are those who relied on position-one rankings without building citation presence inside the AI layer. Ahrefs analyzed roughly 300,000 keywords and found that position-one rankings lose 34.5% of their CTR when an AI Overview is present on the same query. The May 2026 update compounds that dynamic: cited sources are now more prominent and organic links further below the fold are less visible. Ranking and being cited are no longer the same thing, and the gap between them is widening.
What does the CTR data say about citation presence?
The traffic impact of AI Overviews is now documented across multiple independent datasets. Organic CTR on queries that trigger an AI Overview dropped sharply as the feature expanded: one tracking study measured a fall from 1.76% in June 2024 to 0.61% by September 2025, before a partial recovery to around 2.4% by February 2026 as Google refined how much SERP real estate the box occupies.
The citation gap is the clearest signal in the data. Brands cited inside the AI Overview earn approximately 120% more organic clicks per impression on the same query compared to uncited brands, according to Omnibound's AI Overviews statistics dataset. Being in the box does not just mean AI-attributed traffic. It means more of the remaining organic traffic pool too, because the cited brand name is visible and trusted at the top of the page before any organic result is considered.
The implication is direct: for queries where AI Overviews are present, citation is not a brand-awareness bonus. It is the primary lever controlling how much of the organic traffic pool you capture. Ranking in position one without being cited is increasingly a low-return position. The user sees the AI answer, they see the cited brands, and they make a decision before their eyes reach your blue link.
How should you respond to the AI Overviews expansion?
The five SERP changes from May 2026 point to the same underlying direction Google has been moving for two years: the engine rewards specificity, sourcing, and depth. Every structural change in the May update makes citation identity more visible and source quality more consequential. The response is not a tactical fix. It is a content and measurement program.
The first step is an audit. Run the queries that matter to your category inside Google and record which AI Overviews appear, which brands are cited, and where inside the response each citation lands. The inline citation format introduced in May makes this more legible than it was before. You can now see exactly which claims are attributed to which sources, and you can reverse-engineer the type of content Google is trusting.
The second step is coverage. AI Overviews cite sources that answer specific sub-questions within a topic, not just broad authoritative pieces. A single pillar article rarely earns multiple inline citations across an AI Overview. A library of tightly scoped articles, each answering one question with a real data point or a precise explanation, creates many more citation surfaces across the full question universe for your category. Publishing scale and topical specificity are the structural advantages that translate into citation share.
The third step is community content. The Expert Advice block is a net-new citation surface and it is not indexed through traditional web crawling in the same way. Direct participation in forums, substantive answers in communities your buyers use, and detailed social posts now have a pathway into the AI Overview that runs parallel to editorial SEO. Ignoring this surface means ceding it to competitors who are already present.
The fourth step is measurement. CTR data from traditional rank-tracking tools will not tell you whether you are being cited in AI Overviews, let alone across ChatGPT, Perplexity, and Gemini. Citation Share, the percentage of relevant AI answers in your category that cite your brand, is the metric that maps to this surface. Tracking it consistently across all AI engines gives you a complete picture of your answer presence and shows you where competitors are winning ground you cannot see in a rank tracker.
Source: The Stacc / SeoProfy AI Overviews Statistics 2026, 2026-07-01
Key takeaways
- On May 6, 2026, Google deployed five structural changes to AI Overviews: inline citations, hover previews, an Expert Advice block, subscription labels, and an Explore New Angles section.
- AI Overviews now appear on approximately 48% of US Google queries, up from 6.49% in January 2025, and reach 2.5 billion users across 200 countries globally.
- The inline citation change makes brand identity visible next to specific claims before the user clicks, raising the brand-impression value of every citation earned.
- The Expert Advice block is a net-new citation surface for community content: Reddit threads, forum posts, and social contributions can now surface inside AI Overviews without a traditional editorial link.
- Brands cited in AI Overviews earn roughly 120% more organic clicks per impression versus uncited brands on the same query.
- Citation Share across AI engines is the metric that maps to this surface. Traditional rank tracking alone will not show you whether you are winning or losing in the AI answer layer.
Omnicite Editorial. "Google AI Overviews Update: What Changed and What to Do" The Citation Report, Omnicite. https://omnicite.co/blog/ai-overviews-expansion-response/
Sources
May 6, 2026 AI Overviews update: five structural changes including inline citations, hover previews, Expert Advice block, subscription labels, and Explore New Angles NetConnect Digital, 2026-05-06
AI Overviews reach more than 2.5 billion users per month across 200 countries and territories; Gemini 3.5 Flash is the new default model in AI Mode globally Google Blog, 2026-05-20
AI Overviews available in more than 200 countries and territories and more than 40 languages as of May 2025, with support added for Arabic, Chinese, Malay, and Urdu Google Blog, 2025-05-01
AI Overviews appeared in 6.49% of US queries in January 2025, 24.61% in July 2025, and approximately 48% of US searches by mid-2026 The Stacc, 2026-07-01
In January 2025, 91.3% of AIO-triggering queries were informational; by October 2025 that fell to 57.1%, with commercial queries reaching approximately 19% SeoProfy, 2026-01-01
Brands cited in AI Overviews earn approximately 120% more organic clicks per impression versus uncited brands on the same query Omnibound, 2026-01-01
Position-one rankings lose 34.5% CTR when AI Overviews are present, based on a study of approximately 300,000 keywords Ahrefs, 2025-06-01
Frequently asked questions
What is the biggest change in the May 2026 Google AI Overviews update?
Citations moved from a grouped carousel at the bottom of the AI Overview box to inline positions next to the specific text they support. Source identity is now visible before the user clicks, making the brand behind each claim legible directly on the page.
How many Google searches now trigger an AI Overview?
As of mid-2026, AI Overviews appear in approximately 48% of US Google searches, up from 6.49% in January 2025. Coverage is highest on informational queries but has expanded significantly into commercial and navigational query types.
Does being cited in AI Overviews affect organic click-through rates?
Yes. Brands cited inside AI Overviews earn roughly 120% more organic clicks per impression on the same query compared to uncited brands. Being in the box matters not just for AI-attributed traffic but for the remaining organic traffic pool too, because cited brand names are visible at the top of the page before any organic result is reached.
What is the Expert Advice block in AI Overviews?
The Expert Advice block, introduced in the May 6, 2026 update, surfaces first-hand perspectives from Reddit, forums, and social media inside the AI Overview. Creator names and handles are displayed. It is a separate citation surface from the main body of the AI response and creates a new pathway into the AI answer layer for brands with community presence.
How should a brand respond to the AI Overviews expansion?
Start with an audit: run the queries that matter to your category and record which brands are cited and where. Build coverage with tightly scoped articles that each answer one specific question with a real data point. Participate in communities to target the Expert Advice block. Then measure Citation Share across AI engines rather than relying on traditional rank tracking.
Does the AI Overviews update affect all industries equally?
No. Coverage varies significantly by query type and vertical. Informational queries remain the most common trigger, but commercial and navigational query coverage is growing fastest. Categories such as B2B SaaS, local services, health, and finance see high AI Overview rates on decision-stage queries where the stakes for citation presence are highest.