Can AI crawlers read JavaScript-rendered pages?

Updated 10 Nov 2026 · by Venu Gopal Reddy Muvva, founder of BrandInAIAnswers · official sources checked 10 Nov 2026

Plan as if AI crawlers see only the raw HTML your server sends: OpenAI's, Anthropic's and Perplexity's crawler pages do not say they run JavaScript. If your prices, product names or articles appear only after scripts run, those facts may be invisible to AI answers; server-side rendering or pre-rendering fixes it.

Test what a crawler receives

curl -s -A "Mozilla/5.0 (compatible; OAI-SearchBot/1.0)" https://example.com/pricing \
  | sed -e 's/<[^>]*>/ /g' | tr -s ' ' | head -c 2000

If the output does not contain your plan names and prices, a crawler that does not render JavaScript will not see them either. Google is different: its Search crawler renders JavaScript, which you can confirm with URL Inspection.

Pages that most often break

PageTypical causeFix
PricingPrices loaded from an API after page loadRender prices on the server
DocsClient-only docs appStatic site generation or pre-rendering
Product listingsInfinite scroll with no HTML fallbackPaginated HTML pages with links
FAQsAccordion content injected on clickKeep answers in the HTML, hide with CSS
ReviewsThird-party widget in an iframeSummarise key review facts in your own text

Fix options, cheapest first

  • Move the key facts (name, price, plan limits, address) into static HTML on the page.
  • Turn on server-side rendering or static generation in your framework (Next.js, Nuxt, Astro, SvelteKit all support it).
  • Use a pre-rendering service for old single-page apps.
  • Publish Markdown versions of key pages, linked from llms.txt.

A quick audit for your top ten pages

List the ten pages buyers visit most (home, pricing, top product and use-case pages, contact). For each, run the curl test above and note whether the page title, the main heading, the key facts (price, plan names, address) and at least a few paragraphs of copy appear in the raw HTML.

Any page that fails is a candidate for server rendering first, because those are the pages AI answers most often need. Re-run the test after each fix; it takes seconds and catches regressions when someone adds a new client-side widget.

Questions

Does Googlebot render JavaScript?

Yes, Google Search renders pages, and AI Overviews use the Search index. Other AI crawlers are the concern.

How much text should the raw HTML have?

All the facts a buyer needs. A page with fewer than 100 words in raw HTML and lots of script is a warning sign.

Is hiding text with CSS a problem?

Content collapsed in an accordion but present in the HTML is fine and readable.

See which AI answers name you today

BrandInAIAnswers's free audit checks whether AI crawlers can read your site and asks an AI with web search the questions your buyers ask, then shows who gets named. The first 14 days of the full app are free.

Sources

Related

Can AI crawlers read JavaScript? Test your pages