AI Overviews stay scarce in health, finance and legal search results

AI Overviews are not spreading evenly across search. In health, finance, and legal queries, Google is running a much tighter trust filter, and the brands that show up have to clear an authority test before they clear an SEO test.

A trust hierarchy, not a universal rollout

This is the core mistake people make with AI search: they treat it like a single new layer that behaves the same way everywhere. It does not. Search Engine Land’s analysis frames health, finance, and legal as YMYL territory, short for your money or your life, where a bad answer can affect well-being, finances, or safety, so Google applies extra caution before showing an AI Overview.

That caution changes the visibility game. In lower-stakes topics, AI Overviews can appear much more freely, but in YMYL categories the feature may not appear at all, or it may only surface when Google finds sources it trusts enough to lean on. In practice, that means raw page-one placement is not the whole story anymore. A brand can rank well in organic search and still fail to clear the trust threshold for AI visibility.

What Google says the system is doing

Google’s own explanation matches that selective approach. The company says AI Overviews appear when its systems determine generative AI would be especially helpful, and that the feature is still expanding gradually to more users, languages, and regions. Google also says AI Overviews are shown only when they add value beyond classic Search, not as a replacement for it.

The mechanics matter too. Google says AI Overviews and AI Mode can use query fan-out, which means issuing multiple related searches across subtopics and data sources before assembling an answer. Search Central also says there are no special optimizations required beyond standard SEO best practices and helpful, reliable, people-first content. In other words, there is no secret AI checkbox to flip. The same fundamentals still matter, but Google is judging them through a more selective lens.

Google’s own help pages also carry a blunt warning: AI Overviews can make mistakes, and users should double-check important information in more than one place. That warning is especially telling in YMYL categories, because it reinforces the idea that AI search is not meant to be an unquestioned authority.

The numbers show the gap is real

The clearest proof comes from Ahrefs’ scale. Its analysis of 146 million SERPs found AI Overviews on 21% of keywords overall. That sounds broad, until you break it into categories. Medical YMYL queries triggered an AI Overview 44.1% of the time, while news queries triggered one only 6% of the time.

That spread tells you two things at once. First, Google is not using one flat rule for all queries. Second, category sensitivity is real enough to swing AI visibility by dozens of percentage points. If you work in health, finance, or legal search, you are not competing only against other pages. You are competing against the system’s own caution.

Health is the clearest warning shot

Health search is where the trust hierarchy becomes easiest to see, and most uncomfortable. Search Engine Land reported on a SE Ranking study of 50,807 health-related searches in Germany, and the results were messy in exactly the way you would expect from a fast-moving AI layer. Nearly two-thirds of the Google AI Overview citations came from sources without strong medical or evidence-based safeguards.

One detail stands out even more: YouTube was the single most cited source for health-related AI Overviews. That is a reminder that citation volume and authority are not the same thing, especially in a space where the quality of the source should matter as much as the answer itself. The same study found only 36% of AI-cited pages appeared in Google’s top 10 organic results, which means AI citations are not just a mirror of classic rankings.

For health brands, that is the big lesson. AI visibility is not simply about being high enough on the page. It is about whether your source looks safe, specific, and credible enough for Google to trust in a category where the wrong answer can do real harm.

What finance and legal brands should read into it

Finance and legal search follow the same logic, even if the exact mix of citations is different. These are also YMYL categories, so Google is not just looking for relevance. It is looking for signs that the source is authoritative enough to carry risk-sensitive information without drifting into noise, half-truths, or overconfident summaries.

That is why the old SEO habit of treating every query the same way falls apart here. A generic landing page with broad keyword targeting is far less likely to matter than a page that shows clear expertise, cites dependable sources, and reads like it was built by people who actually understand the subject. In YMYL, the trust signal is part of the ranking signal, even if Google keeps saying the mechanics are not special in the usual optimization sense.

Search Engine Land’s framing, along with coverage from contributors like Zoe Ashbridge, Rebecca Bridge, Ryan Law, Xibeijia Guan, Barry Schwartz, Hema Budaraju, and Duncan Osborn, points to the same practical conclusion: these sectors do not play by the same visibility rules as entertainment, lifestyle, or general how-to content.

What actually helps a brand clear the trust gate

The playbook is less glamorous than many AI search pitches, but it is more durable.

• Publish under clearly identifiable experts with real credentials.
• Use sources that are easy to verify and strong enough to withstand scrutiny.
• Make the page’s purpose obvious, especially when the topic involves money, health, or legal consequences.
• Keep content current and specific, not padded with generic advice.
• Build for helpfulness first, because Google says there are no special AI-only tricks beyond standard SEO best practices.

That lines up with Google’s broader direction too. In 2026, the company said it is updating AI Search to help users find original content and trusted sources more easily, including more direct links and source previews. That is a loud signal that source authority is becoming more visible inside the AI experience itself.

The bottom line is simple: AI search is not one universal layer. It is a category-sensitive system, and YMYL brands have to earn visibility through trust first, optimization second. In health, finance, and legal search, the brands that win are the ones Google is willing to trust when the stakes are highest.