Tweets: A 41-Day Test Just Showed AI Crawlers Can't See Your JavaScript Links

1. A 41-day test split one site's internal links in half: half in plain HTML, half injected by JavaScript. GPTBot, ClaudeBot, and Bingbot found zero of the JavaScript-linked pages. Not slowly. Zero. [LINK]

2. Googlebot was the only crawler in the test that even attempted to render JavaScript, and it still only reached about 2% of the JS-linked pages. Every other major bot reached none of them. [LINK]

3. When the site owner switched those links back to plain HTML, GPTBot discovered roughly 250 new pages quickly, and Bingbot found about 212. Same content. Different discovery path. Completely different result. [LINK]

4. A separate analysis of over 500 million crawler fetches confirms it: none of the major AI bots, including OpenAI's and Anthropic's, execute JavaScript at all. They fetch raw HTML and move on. [LINK]

5. You can refresh a page all year and it won't matter if the internal link pointing to it only exists after JavaScript runs. AI crawlers never get that far to see the improvement. [LINK]

6. A related-posts widget or a JavaScript-built nav menu can quietly wall off an entire section of a site from every AI crawler at once. Most teams have never checked. [LINK]

7. Fastest way to check: pull the raw HTML for a page with curl, not a browser. If your links and content aren't in that response, GPTBot and ClaudeBot will never see them either. Full breakdown here. [LINK]