Does llms.txt actually help you show up in AI search?

llms.txt is a plain-text file some sites publish at their root to summarize their content for AI systems. Adoption picked up fast through 2026 on the theory that AI platforms would reward it. The evidence says otherwise.

97% of llms.txt files studied across 137,000 domains received zero requests of any kind in May 2026, according to Ahrefs. Google's own Search Central documentation says the file has no effect on Search or AI Overviews. The one place llms.txt may genuinely matter is different from where most people think it does.

What llms.txt actually is

llms.txt is a single Markdown file placed at a website's root, similar in spirit to a sitemap. It was proposed in 2024 by Jeremy Howard, co-founder of Answer.AI, as a way for AI systems and agents to quickly understand what a site is about without crawling every page. It is not a directive file like robots.txt. It doesn't block or control anything. It's a summary, and it's only useful if something actually reads it.

What Google says, directly

Google's Search Central documentation addresses llms.txt under a section titled "Mythbusting generative AI search: what you don't need to do." The guidance states that sites don't need to create llms.txt files, AI text files, or special markup to appear in Google Search or its generative AI features, because Google Search doesn't use them. It adds that publishing one will neither harm nor help a site's visibility or rankings, since Google Search ignores them entirely. That's about as direct as Google's documentation gets.

Google Search ignores llms.txt. It says so in writing.

What the data shows

Ahrefs analyzed 137,210 domains in May 2026 and found that 28% publish an llms.txt file, but 97% of those files received zero requests, no bots, no humans, the entire month. Of the small share that did get traffic, 96% of requests came from bots rather than people.

Breaking those bot requests down further tells the more useful story. AI retrieval bots, the ones that actually answer live search queries in ChatGPT, Perplexity, and similar tools, accounted for just 1.1% of all requests. Slack's link-preview bot alone fetched llms.txt files more often than Perplexity's search crawler did. The single largest category of requests, 21.7%, came from SEO audit tools, not AI systems at all.

Where it might actually matter

The data pointed to one real exception: AI coding agents. Anthropic's Claude Code outfetched every AI search bot, AI assistant, and AI training crawler in the study. Separately, Chrome's Lighthouse tool added an llms.txt check to its Agentic Browsing audits in 2026, aimed at AI agents that browse and act on a user's behalf, like filling out a form or comparing products, not at search visibility. Chrome's own documentation for the audit treats a missing file as "not applicable," since providing one is optional. Google's John Mueller has described llms.txt as a temporary stopgap for AI coding tools parsing documentation, not something most business sites need to worry about.

Should you still create one?

There's no harm in it. It costs almost nothing to write, and some site builders now generate one automatically. But it shouldn't take time or budget away from what's actually proven to move AI citations: a complete, schema-marked website, sustained review velocity and third-party corroboration, and content built around the specific conditions, services, and questions customers are actually asking AI about. Those three levers are backed by real citation-testing data. llms.txt, for now, is not.

The takeaways

• 97% of llms.txt files across 137,000 domains received zero requests in May 2026, from bots or humans.
• Google's Search Central documentation states directly that llms.txt has no effect on Search rankings or AI Overviews.
• AI retrieval bots behind ChatGPT and Perplexity search answers made up only 1.1% of the small share of requests that did occur.
• The closest thing to a real use case today is AI coding agents and browser-agent tools, not AI search visibility.
• What reliably drives AI citations hasn't changed: verified business data, third-party corroboration, and specific, citable content.