
Short Answer: Can AI search read JavaScript websites?
Only if the important content is in the initial HTML, or the crawler runs JavaScript. Google Search generally renders JavaScript before indexing a page. Neither OpenAI nor Anthropic documents that its crawlers do, and the public tests we found suggest many AI crawlers read the server-delivered HTML only. Content that appears after a script runs can be missed.
Key Takeaways
- Google says it runs JavaScript with a headless Chromium browser, but rendering is queued and is not instant. (Google)
- OpenAI and Anthropic document their crawlers’ purposes but say nothing about executing JavaScript. (OpenAI, Anthropic)
- In Vercel’s December 2024 log analysis, the GPTBot, OAI-SearchBot, ChatGPT-User, ClaudeBot, and PerplexityBot crawlers did not render JavaScript. That is dated, single-source evidence, not a rule. (Vercel)
- You can check any page in minutes: compare View Source with the live page, fetch the raw HTML with curl, and reload with JavaScript disabled.
- If a fact matters for search or AI visibility, put it in the HTML the server sends first. Structured data cannot stand in for missing visible text.
That gap, between what the server sends first and what a browser builds afterward, drives everything below: how rendering works, how AI crawlers differ from Google, and how to test your own pages before involving a developer.
This article is about rendering only. It does not cover robots.txt rules, licensing, or a full crawler-access audit. For those, see our AI crawler user-agent list.
Client-Side Rendering, Server-Side Rendering, and Hydration in Plain Terms
Think of a webpage as a package being delivered.
- Server-side rendering (SSR): The box arrives already packed. The server builds the full HTML, including headings and text, before sending it to the browser or crawler.
- Client-side rendering (CSR): The box arrives empty, with a set of instructions (JavaScript) and tools. The browser has to build the contents after it arrives.
- Static site generation (SSG): The box was packed in advance, during a build step, and every visitor gets the same pre-filled version.
- Hydration: A hybrid approach. The box arrives pre-filled with basic content, and JavaScript later adds interactive behavior on top of it.
What the server sends vs. what the browser builds
For a heavily client-rendered site, the server response often contains little more than an empty container such as <div id="root"></div> plus script tags. JavaScript running in the browser then inserts the article text, product details, and navigation links.
A crawler that does not execute JavaScript sees only that empty shell. A crawler that does execute it, or a server that pre-renders the page, sees the finished content.
Does Google Search Render JavaScript?
Yes, generally. Google’s JavaScript documentation describes three phases: crawling, rendering, and indexing. Googlebot fetches the page, queues it for rendering, and a headless Chromium browser then runs the page’s JavaScript. Google indexes the rendered HTML and uses it to discover more links.
Rendering is not instant, though. Google says a page can wait in the render queue for a few seconds or longer, and that rendering happens when its resources allow. It also says Google Search will not render JavaScript from files that robots.txt blocks, so a page whose key scripts are blocked can render incomplete or empty.
Do AI Search Crawlers Render JavaScript the Same Way?
Not necessarily, and behavior is not uniform across AI search. Two different mechanisms are worth separating.
Search-index-based retrieval
Some AI features draw on an existing search index instead of rendering pages themselves. Google, for example, says a page must be indexed and eligible to appear in Search with a snippet to be shown as a supporting link in AI Overviews and AI Mode. For those features, visibility depends more on whether Google already indexed the rendered content, as described above.
Direct-fetch AI crawlers
Other AI systems run their own crawlers. OpenAI documents three: GPTBot, which crawls content that may be used to train its models; OAI-SearchBot, used to surface sites in ChatGPT search; and ChatGPT-User, which visits pages for certain user-initiated actions. Anthropic documents ClaudeBot for training data, Claude-User for user requests, and Claude-SearchBot for improving search results. Neither company’s page says whether these bots execute JavaScript. For a closer look at the OpenAI bots, see our guide to GPTBot, OAI-SearchBot, and ChatGPT-User.
The best public evidence we found comes from outside the vendors. In a December 2024 log analysis, Vercel and MERJ reported that none of the major AI crawlers they tested rendered JavaScript, including GPTBot, OAI-SearchBot, ChatGPT-User, ClaudeBot, and PerplexityBot. They saw OpenAI’s crawlers request JavaScript files in 11.5% of fetches and Claude’s in 23.84%, without executing them. Gemini, which uses Googlebot’s infrastructure, and Apple’s AppleBot did render. The data came mainly from nextjs.org and Vercel’s network, with two job-board sites added for cross-checking.
Advanced Web Ranking’s review of AI crawler tests draws the same broad conclusion. It does not run a new test. It summarizes the Vercel data and other third-party experiments, including a searchVIU test in which Gemini was the only system that picked up a price inserted purely by JavaScript during a live fetch.
Treat all of this as a snapshot, not a rule. The main log study is almost two years old, covers a limited set of sites, and describes crawler behavior at the time. Crawlers can change without notice, and the vendors do not commit to a rendering behavior. The safest approach is to test your own pages and check your own server logs instead of assuming either extreme.
Why JavaScript Can Hide Content From AI Crawlers
Content commonly goes missing from crawlers that do not run JavaScript for a few recurring reasons:
- The initial HTML contains only an empty app shell
- Article or product text loads from an API call after the page opens
- Links depend on click handlers instead of standard
<a href>tags - Content appears only after scrolling, a click, or a login
- JavaScript files, APIs, or third-party scripts are blocked from loading
- The page returns an error, a redirect loop, or a slow response that interrupts loading
Any of these can leave a crawler with little to analyze, even if the page looks complete to a human visitor. The link point applies to Google too: it says it can only discover links that are <a> elements with an href.
How to Check If Your Content Is Visible Before JavaScript Runs
Step 1: View Source vs. Inspect Element
Right-click a page and choose “View Page Source.” This shows what the server actually delivered. The Inspect Element panel, by contrast, shows the page after JavaScript has modified it. If visible content appears in Inspect Element but not in View Source, a script added it.
Step 2: Fetch the raw HTML with a simple request
A basic command-line request shows the server’s first response without any browser rendering:
curl -L https://example.com/page
Hypothetical example: if example.com is a product page, search the returned text for the product name or price. If it is missing from this output but visible in a browser, JavaScript is adding it afterward.
Step 3: Disable JavaScript in Chrome DevTools
Open DevTools, press Cmd+Shift+P (Ctrl+Shift+P on Windows or Linux) to open the Command Menu, type “javascript,” and choose “Disable JavaScript,” then reload. Chrome’s documentation notes this applies only to that tab while DevTools is open. Check whether the main text, headings, and navigation links still appear. This is a rough approximation of a non-rendering crawler, not an exact match for any specific bot.
Step 4: Compare an HTML-only crawl with a rendered crawl
Crawling tools that support both HTML-only and JavaScript-rendered modes can run the same site twice and show the difference in word count, headings, links, and structured data. A large gap between the two is a warning sign.
Step 5: Use Google’s own rendering tools
Google’s URL Inspection tool lets you test a live URL and see how Google fetched it, including a screenshot, the returned HTML, and the page resources. Google’s troubleshooting guide says the tool also shows JavaScript console output and exceptions and the rendered DOM. This diagnoses Google Search specifically. It does not prove how any AI crawler will handle the same page.
What to Look for in the Raw HTML
Once you have the server-delivered HTML, check for these elements:
- Main heading (H1) and subheadings
- Full body text, not a summary or placeholder
- Product details such as name, description, and price, where relevant
- Title tag and meta description
- Standard internal links using
<a href> - Structured data in JSON-LD format, if used
If several of these are missing from the raw HTML but present in the browser, that page depends on JavaScript to deliver its core content.
JavaScript Content Audit Template
Use this table to self-audit key page types before involving a developer. It is blank by design: fill in your own results.
| Page type | Main text in raw HTML? | Headings in raw HTML? | Links use <a href>? | Price/details in raw HTML? | Structured data present? | Risk level |
|---|---|---|---|---|---|---|
| Homepage | Yes / No | Yes / No | Yes / No | Yes / No | Yes / No | Low / Med / High |
| Product page | Yes / No | Yes / No | Yes / No | Yes / No | Yes / No | Low / Med / High |
| Blog or article page | Yes / No | Yes / No | Yes / No | Yes / No | Yes / No | Low / Med / High |
| Category or listing page | Yes / No | Yes / No | Yes / No | Yes / No | Yes / No | Low / Med / High |
Mark each column yes or no using Steps 1 and 2 above, then flag any row with multiple “no” answers as higher risk for JavaScript-dependent content.
Fixing Gaps Without Removing JavaScript
JavaScript does not need to be removed to fix this problem. Options include:
- Server-side rendering or static generation so the server delivers finished HTML rather than an empty shell
- Hydration, where usable HTML arrives first and JavaScript adds interactivity afterward
- Crawlable links, using standard
<a href>markup instead of click-only navigation - Structured data (JSON-LD) that describes content already present in the visible HTML, as reinforcement rather than a replacement for real text. Our look at whether schema markup helps AI search visibility covers why.
Google calls dynamic rendering, which serves bots a separate pre-rendered version, “a workaround and not a recommended solution,” and points to server-side rendering, static rendering, or hydration instead. Those approaches also serve crawlers that do not run scripts. Our comparison of HTML and Markdown for AI crawlers makes a related point: what the crawler can read in the page is what counts.
Learn More About AEO and AI Marketing at Prompt Insider
Since launching earlier this year, Prompt Insider has become a leading authority on AI marketing, Answer Engine Optimization (AEO), large language models, AI search, AI news, and the evolving future of digital discovery. As AEO becomes one of the hottest topics in marketing, Prompt Insider is helping define the conversation around how brands improve visibility, adapt their content strategies, and stay competitive in an increasingly AI-driven search environment.
Prompt Insider is the go-to resource for answer engine optimization, AI marketing, and AI search. Start with our core guides at thepromptinsider.com:
- What Is AEO? Answer Engine Optimization Explained
- AEO vs. SEO vs. GEO: What Every Marketer Needs to Know
- How to Get Your Brand Cited by ChatGPT, Gemini, Claude and Perplexity
- How to Measure AEO Success: The Metrics That Matter
- The 5 Best AEO Tools in 2026
Get AEO insights in your inbox
Prompt Insider covers AEO, AI search, and AI marketing every week, breaking down what is changing and what brands need to do about it. Sign up for our emails at thepromptinsider.com to get it first.
Frequently Asked Questions
Can ChatGPT crawl JavaScript websites?
There is no single documented answer. OpenAI lists GPTBot, OAI-SearchBot, and ChatGPT-User, but its crawler page does not say whether they execute JavaScript. In Vercel’s late-2024 logs, OpenAI’s crawlers requested JavaScript files without running them. Test your own pages with the methods above instead of assuming a blanket answer.
Can Google read React, Vue, or Angular websites?
Usually, yes, based on Google’s documented crawl, render, and index process. Rendering can still be delayed, and Google will not render JavaScript from files that robots.txt blocks, so test key pages with the URL Inspection tool even on JavaScript frameworks.
How do I check whether my content is readable without JavaScript?
Compare the raw HTML to the rendered page: use View Source, a curl request, and a reload with JavaScript disabled. Server logs can also show which bots request your pages. This confirms whether content is technically readable, not whether it will be selected, ranked, or cited, which depends on factors beyond rendering.
Does robots.txt stop AI crawlers from reading JavaScript content?
Robots.txt signals access preferences to crawlers that choose to follow it. It is not a security or access-control mechanism, and it does not change whether a crawler executes JavaScript. Blocking your script files can still stop Google from rendering a page correctly.
Does adding JSON-LD structured data fix a JavaScript rendering problem?
No. Structured data should describe content that already exists as readable HTML. It supplements visible text; it does not substitute for content that only appears after a script runs.
Bottom Line
If a fact matters for search visibility or AI citations, make sure it exists as plain, crawlable HTML in the initial server response. Google generally renders JavaScript, but rendering is queued and can fail. For AI crawlers, the vendors say nothing and the public tests we found point toward HTML-only reading, so an HTML-first page is the more reliable bet across search engines and AI systems.
Fixing rendering removes a technical barrier. It does not guarantee citations or rankings, which also depend on relevance, authority, freshness, and how each AI system selects sources. For the wider picture, start with our guide to what AEO is and our explainer on how AI search retrieval works.
Sources: Google Search Central: JavaScript SEO basics, Google Search Central: Fix Search-related JavaScript problems, Google Search Central: Dynamic rendering, Google Search Central: AI features and your website, Google Search Console Help: URL Inspection tool, OpenAI: Overview of OpenAI crawlers, Anthropic: Does Anthropic crawl data from the web?, Vercel: The rise of the AI crawler, Advanced Web Ranking: What AI search crawlers read, Chrome for Developers: Disable JavaScript. Study findings are the publishers’ own and are not independently verified. Crawler behavior and vendor documentation change over time. The curl example is hypothetical. We reviewed these sources on November 3, 2026.
About the author
Kai Williams
Kai Williams has been in marketing for years, with a long background in SEO before AEO had a name. He stepped into Answer Engine Optimization the moment AI started reshaping how people search, and has been tracking the shift ever since. At Prompt Insider, he covers AEO, AI marketing, and the future of search, breaking down what is changing and what brands need to do about it.


