AI crawlers
Can ChatGPT actually read your website?
AI crawlers do not run JavaScript. If your content only appears after scripts load, ChatGPT never sees it. Here is how to check in two minutes.
Published July 12, 2026
Short answer: only if your content is in the raw HTML. ChatGPT, Claude, and Perplexity send out crawlers to read websites. Those crawlers do not run JavaScript. An analysis of about 500 million GPTBot fetches, as of mid-2026, found zero JavaScript execution.
That matters more than it sounds. Many modern websites are built so that the server sends a nearly empty page, and JavaScript fills in the text, prices, and buttons afterwards. A person with a browser sees everything. A crawler sees the empty shell.
Why this costs you money
More buying decisions now start in a chat window. When someone asks an AI assistant "what does this company do" or "how much does this cost", the answer comes from whatever the crawler managed to read. If your pricing loads through JavaScript, the assistant either says nothing about you or guesses. Either way, the person asking moves on to a competitor whose site it could read.
And these are good visitors to lose. People who arrive from an AI answer already know what they are looking for, so far more of them become customers than typical search traffic.
How to check your own site
You do not need any tools for a first look. Two ways:
The view-source test. Open your most important page. Right-click and choose "View page source". Now search that code for your product name, your price, and your main heading. Everything you find there exists for AI. Everything you cannot find does not.
The curl test. If you are comfortable with a terminal, fetch your page the way GPTBot does:
curl -A "GPTBot" https://your-site.com | grep -i "your product name"
No match means the crawler never receives that text.
We run a stricter version of this comparison in our own tool. We fetch the page twice, once as a crawler and once as a person, and measure the difference. The methodology is public if you want the details.
The fix, in order of effort
| Fix | Effort | What it does |
|---|---|---|
| Server-side rendering | Medium | Your server sends the full text in the HTML |
| Static generation | Medium | Pages are built as complete HTML files ahead of time |
| Prerendering service | Low | A service serves a rendered copy to crawlers |
If your site runs on a framework like Next.js or Nuxt, server-side rendering is usually a configuration change, not a rebuild. If you use a website builder, check whether it renders content on the server. Most of the well-known ones do, but their add-on widgets often do not.
Start with the one page that earns you money. Fix that first, then work outwards.
The fastest way to know where you stand is to measure it.
Common questions
- Do AI crawlers like GPTBot run JavaScript?
- No. An analysis of about 500 million GPTBot fetches, as of mid-2026, found zero JavaScript execution. GPTBot, ClaudeBot, and PerplexityBot read the raw HTML your server sends and nothing more.
- How do I check what an AI crawler sees on my page?
- Open your page, right-click, and choose "View page source". That raw HTML is what the crawler gets. If your product names, prices, or contact details are missing from it, they are missing for the AI too.
- My site looks fine in Google. Does this still matter?
- Yes. Google renders JavaScript for its classic index, so a site can rank in Google and still be unreadable to AI assistants. The two systems read your site in different ways.
Related reading
See what AI crawlers see on your site
The free Siteiz scan reads one page of your site the way an AI crawler does and grades it A to F. It takes about 30 seconds.