For most of 2026, this has been one of the most argued-over questions in SEO circles — and as of a few weeks ago, it’s not really a debate anymore. Google settled it directly. If you’ve been told adding an llms.txt file to your site will boost your visibility in Gemini or AI Overviews, here’s what the evidence — and now Google itself — actually says.
The Short Answer
No. llms.txt does not currently help you rank in Google Search, AI Overviews, or Gemini. On June 15, 2026, Google updated its official Search Central documentation with a section specifically titled “Clarifying guidance on llms.txt files,” stating in unusually direct language that the file has no effect — positive or negative — on Search visibility, because Google Search does not use it at all.
This wasn’t a policy change. It was Google confirming what independent research had already been pointing to for months.
What llms.txt Was Supposed to Do
The idea behind llms.txt is reasonable on paper. It’s a plain-text file placed in your site’s root directory, meant to give AI language models a condensed, curated summary of your most important content — similar in spirit to how robots.txt tells crawlers what they can and can’t access, except llms.txt was designed to point AI tools toward what’s worth reading rather than restrict anything.
For documentation-heavy platforms, developer tools, and knowledge-base-driven sites, that’s a genuinely useful concept. The problem is what happened next: the narrative quietly shifted from “llms.txt helps AI systems locate your best resources” to “llms.txt helps your site rank and get cited in AI results.” Those are two very different claims, and only the first one has ever had real support behind it.
What the Actual Data Shows
Before Google’s clarification, a handful of independent studies had already tested the ranking claim directly, and none of them found evidence to support it:
- One analysis tracking roughly 94,000 URLs across more than 11,000 AI responses — testing ChatGPT, Gemini, Claude, Grok, and Perplexity — found no measurable citation advantage for sites with an llms.txt file.
- A separate study covering around 300,000 domains found that only about 1 in 10 had implemented llms.txt at all, and found no correlation between having the file and how often a domain got cited in AI-generated answers.
- Traffic analysis of major AI crawlers — GPTBot, ClaudeBot, PerplexityBot, OAI-SearchBot, and Google-Extended — showed they overwhelmingly skip requesting /llms.txt entirely and crawl your normal HTML pages directly instead.
That last point is the most telling. If the AI systems the file is supposedly written for are rarely even fetching it, it’s difficult to see how it could be influencing what they cite.
Why Google’s Clarification Matters More Than the Studies
Independent research is useful, but it can always be argued away as incomplete or outdated. Google removing the ambiguity directly closes that door. The June 15 update specifically told site owners they don’t need special machine-readable files to appear in Search, and reaffirmed something worth sitting with: the same foundational principles that drive traditional rankings are also what drive visibility in AI Overviews and AI Mode. There’s no separate, secret ruleset — no shortcut file that skips the actual work.
This lines up with everything we’ve covered elsewhere in this series. In our explanation of how Gemini’s AI search actually works, the deciding factors are entity trust, content depth, and technical accessibility — not the presence of a supplementary text file. Google’s statement is essentially a confirmation of that same conclusion from the source itself.
So Is llms.txt Completely Useless?
Not entirely — it just isn’t an SEO or GEO lever. There’s a legitimate, narrower use case worth understanding: some describe it as a “business-to-agent” file rather than a ranking file. If you run a platform with an API, extensive documentation, or a product AI agents might need to interact with programmatically — think developer tools, SaaS platforms with public APIs, or knowledge bases meant for AI-assisted workflows — llms.txt can function as a structured pointer for agents that specifically look for it, separate from anything related to search visibility.
For the average business trying to get found and recommended in Gemini or AI Overviews, though, that use case doesn’t apply, and building one in the hope it improves citations is time better spent elsewhere.
What Actually Moves the Needle Instead
If llms.txt isn’t the lever, the honest answer is that the fundamentals haven’t changed as much as the hype around AI search tools suggests. The signals that build trust with traditional search are largely the same ones that earn citations from Gemini, ChatGPT, and Perplexity:
- Clear, direct content that states its answer plainly near the top of the page, rather than burying it under paragraphs of preamble.
- Structured data and schema markup, which reduces ambiguity about who you are and what your content covers — we go deeper into exactly which schema types matter most in our entity-graphing guide.
- Genuine information gain — content that says something the rest of the web hasn’t already said, which we break down fully in our piece on why AI models reward original content.
Real third-party citations and mentions, building the kind of cross-web consensus that no single file on your own server can fake.
This is precisely the sequence of work covered in our Gemini SEO services — technical foundations, entity clarity, and original content, in that order. None of it involves a shortcut file, because there isn’t one.
The Bigger Lesson Here
The llms.txt story is a useful case study in how quickly AI-SEO hype can outrun the evidence. A reasonable, narrow-purpose file got reframed as a ranking hack almost overnight, and it took Google directly stepping in to settle the confusion. It’s worth remembering the next time a new “AI SEO trick” starts circulating — check whether there’s actual data behind it before rebuilding your priorities around it.
Frequently Asked Questions
Should I remove my llms.txt file if I already have one? No need to. Google has stated it neither helps nor harms your rankings, so keeping it causes no damage — it simply shouldn’t be treated as a priority for AI search visibility.
Does this apply to ChatGPT and Perplexity too, not just Google? The June 2026 clarification specifically addresses Google Search and AI Overviews. Independent studies covering ChatGPT, Gemini, Claude, and Perplexity together also found no measurable citation advantage from having the file.
What should I prioritize instead if I want to show up in AI search results? Focus on crawlable technical foundations, clear and directly-answered content, proper schema markup, and genuine authority signals — the same fundamentals that have always driven real search visibility.
