If you've spent any time in GEO or AI-SEO circles in the past year, you've heard that llms.txt is the new robots.txt — the file every AI-ready website supposedly needs. The reality is messier. Adoption is low, Google has explicitly declined to support it, and most of the traffic it was designed to serve barely uses it. That doesn't mean it's worthless. It means the claims around it need a hard look before you spend engineering time on it.
llms.txt is a plain Markdown file that sits at a site's root, similar in concept to robots.txt or sitemap.xml. Instead of crawl rules, it offers a curated, human-and-machine-readable map of a site's most important content — pricing pages, product documentation, integration lists, proof points — written specifically for AI systems and AI agents to consume quickly.
It's important to be precise about its status: as of early 2026, llms.txt remains a community-driven proposal, not a ratified standard from the IETF or W3C. Anyone can implement it, but no governing body has standardized its format or guaranteed any AI platform will read it.
Two data points matter more than any blog post's opinion on llms.txt. First, adoption: broader crawl-based estimates put the share of websites with no llms.txt file above 96%, consistent with the ~10% adoption figure from more targeted domain scans. Second, and more important, is whether AI systems even request the file when it exists.
That ratio — a few hundred targeted requests against half a billion total AI bot visits — is the clearest signal available that most AI crawlers are not prioritizing llms.txt today. And the most consequential holdout is Google. Google's Gary Illyes confirmed in mid-2025 that Google does not support llms.txt for Search or AI Overviews, and has no plans to change that position. Since Google AI Overviews reach roughly two billion monthly users, that's a significant chunk of AI search volume the file simply doesn't touch.
"llms.txt is the first widely-adopted B2A — business-to-agent — standard, particularly among developer tools and AI coding assistants. It is not, today, a GEO citation lever for ChatGPT, Perplexity, or Google AI Overviews."
The place llms.txt shows real signal isn't consumer-facing AI search — it's the growing category of AI agents acting on behalf of a business user. Sales-research agents increasingly embedded inside CRMs and procurement platforms fetch vendor websites directly to summarize pricing tiers, integration compatibility, and case study relevance for a buyer or rep. A well-structured llms.txt file, pointing precisely at pricing pages, integration docs, and proof points, changes what those agents report back — because it removes the ambiguity of parsing a marketing homepage.
This is a genuinely B2B-relevant use case, and it's the reason llms.txt is gaining traction fastest among developer tools and technical products, where agent-to-business interactions are already common.
| Channel | Does llms.txt help? | What actually drives citation there |
|---|---|---|
| ChatGPT / Perplexity conversational search | No measurable effect | Content quality, schema markup, off-site authority (Reddit, G2, reviews) |
| Google AI Overviews | Confirmed: none | Standard crawling, structured data, indexed content quality |
| Sales-research / procurement agents | Yes, directly | A clean llms.txt pointing at pricing, integrations, case studies |
| AI coding assistants / dev tool discovery | Yes, growing adoption | llms.txt as documentation index for the assistant |
Our position: build one, but budget it correctly. It's a low-cost, low-engineering-time addition — usually a few hours to draft and maintain — and it carries essentially no downside. But it should never come out of the budget or timeline allocated for GEO work that actually moves ChatGPT, Perplexity, or Google AI Overview citations, because it doesn't touch any of those.
Any agency or tool promising that an llms.txt file will meaningfully increase your ChatGPT or Google AI Overview citation rate is overstating what the data currently supports. Google has publicly declined to use it, and even among platforms that might read it, request volume is negligible next to total AI crawler traffic. Treat it as good infrastructure hygiene for the B2A use case, not a GEO strategy on its own.
llms.txt is a real, if early, convention with a genuine use case for B2B companies whose buyers increasingly route research through AI agents embedded in CRMs and procurement software. It is not, despite the marketing around it, a way to move the needle on ChatGPT, Perplexity, or Google AI Overview citations — those depend on the fundamentals: cited, well-structured content, schema markup, and off-site authority. Build the file. Don't confuse it with a GEO strategy.
llms.txt is a plain-text file, similar in spirit to robots.txt, that sits at a website's root and gives AI systems a curated, structured map of a site's most important pages — usually pricing, documentation, integrations, and proof points. It's a community-proposed convention, not an official web standard: as of early 2026 it has not been adopted by the IETF or W3C.
No. Google's Gary Illyes confirmed in mid-2025 that Google does not support llms.txt and has no plans to. Google-Extended and Googlebot continue to rely on standard crawling, robots.txt, and structured data — llms.txt has no effect on Google AI Overviews or organic rankings.
Adoption is low. One study scanning 300,000 domains found roughly a 10% adoption rate, and broader crawls put the share of sites with no llms.txt file above 96%. Separately, an analysis of over 500 million AI bot visits across a 90-day window found only a few hundred requests that targeted an llms.txt file directly — negligible next to total AI crawler traffic.
It's a low-cost, low-risk addition worth doing, but not a GEO priority. The strongest use case is agent-to-business scenarios: sales-research agents inside CRMs and procurement tools that fetch a vendor's site to summarize pricing, integrations, and case studies. A well-structured llms.txt can shape what those agents report back. But it does nothing for ChatGPT, Perplexity, or Google AI Overview citation rates, which depend on content quality, schema markup, and off-site authority — not this file.
We'll audit your site's real AI search visibility gaps — separating the fixes that move citations from the ones that just feel productive.
Book a Free AI Visibility CheckFree — No Commitment
We test 15 of your buyers' top queries across ChatGPT, Perplexity, and Google AI Overviews — and show you exactly where you appear, where competitors are being cited, and what's driving the gap.
Check My AI VisibilityTakes 2 minutes to request. We'll follow up within one business day.