Is AI search stricter about health and finance content?

Google's helpful content guidance says E-E-A-T (experience, expertise, authoritativeness, trust) "isn't a specific ranking factor." It's a concept. Google's systems approximate it through a mix of other signals, and they weight it more heavily on YMYL topics: health, money and safety subjects where bad information can hurt someone.

That weighting carries into Google's AI features because there is no separate path into them. Google's AI features documentation says a page must be indexed and eligible to show in Search with a snippet to appear as a supporting link in AI Overviews or AI Mode. It adds: "There are no additional technical requirements."

So when people say AI search is stricter about health and finance, the accurate version is narrower. Google Search already treats YMYL topics with more caution, and its AI surfaces sit on top of the same Search systems. Google documents no extra AI-only YMYL filter. For the rest of what Google has published, see our digest of Google's AI features documentation.

There is no number to move. Any tool that reports an E-E-A-T score for your pages invented that score. Google's Search Quality Rater Guidelines describe raters evaluating results, not ranking them, and say humans "could never individually rate each page on the open web."

What Google does ask for is ordinary and demanding at once:

• Accuracy. Claims about dosages, fees, rates or eligibility should be correct and current.
• Sourcing. Say where a figure comes from, so a reader can check it.
• People-first purpose. Write for the person deciding, not for a crawler.

Google's AI optimization guide adds that you "don't need to write in a specific way just for generative AI search." No special phrasing, chunking or markup is required.

One on-page habit does have some support. Stating a specific fact plainly, on a page that can be fetched, gives an engine a sentence it can quote. Say you run a hypothetical mortgage brokerage: "Our arrangement fee is $495, paid on completion" is quotable. "Competitive fees" is not. This is no promise of a citation. It just removes ambiguity. Sector notes are in our pages on AI visibility for fintech companies and mental health practices.

Outside Google, the public record is thin. As of October 2026, OpenAI's entire description of how ChatGPT search orders results is that it "ranks search results using multiple factors intended to help users find relevant, reliable information. Placement is not guaranteed." Nothing is enumerated, and no YMYL category is named.

Perplexity and Anthropic document their crawlers and APIs, but none of their sources in our evidence base describe topic-specific trust weighting. That means:

• Don't assume Google's YMYL framing transfers to ChatGPT, Perplexity or Claude.
• Don't assume it doesn't. Nobody outside those companies can see.
• Retrieval is discretionary on every platform, so some answers about health or money come from the model's training data with no live search at all.

Engines also draw on different indexes. A page healthy in Google can be missing from Bing, and the reverse. Our comparison of Bing index vs Google index for AI answers covers which surfaces use which.

Invented credentials

Otterly's 2026 experiment loaded two new domains with roughly 1,000 AI-generated posts each. One showed a fictional wildlife biologist in an author box. Otterly reports that site lasted about 14 days in Google before deindexing, while the site with no author lasted about 90. That is one site per arm, so it does not prove author boxes hurt. It is no evidence that they protect thin content either.

Volume

Google's spam policies define scaled content abuse as many pages generated "for the primary purpose of manipulating search rankings and not helping users." Google also says appropriate use of AI is not against its guidelines. The problem is volume without value, and on YMYL topics that pattern carries more risk.

Wrong answers about you

AI platforms get specific facts wrong about specific brands. For a clinic or a lender, a wrong price, licence or service claim matters more than it would for a hobby blog. If Google shows one, our guide to wrong facts in AI Overviews explains the reporting routes.

Hidden instructions

Injected text aimed at AI crawlers (hidden prompts, cloaking) is illegitimate. Research shows it can sway some search agents once it is retrieved. On health and finance topics it is also a fast way to lose trust and index status.

These are the parts in your control. None guarantees inclusion, because Google says indexing and serving "isn't guaranteed."

1. Index and snippet eligibility. Confirm key pages are indexed and not blocked by noindex, nosnippet or max-snippet:0 left over from a template.
2. The Search Console AI control. Inclusion is the default, but a property can be opted out, including by inheriting from a parent. See Google's AI features control in Search Console.
3. Per-bot access for other engines. Check that robots.txt allows OAI-SearchBot, PerplexityBot and Claude-SearchBot, and that your firewall or CDN does not return 403 to them. Blocking GPTBot only affects training.
4. Facts in the raw HTML. Most non-Google AI crawlers don't execute JavaScript, so prices, fees and service details should be in the HTML before scripts run.
5. What each engine actually says. Ask the buyer questions your customers ask, engine by engine, and compare the answers with your own facts. The free checker on this site runs up to three such questions as a snapshot.