Google's John Mueller Debunks LLMs-author.txt as Useless for SEO Identity Problems

A Reddit user's attempt to solve a name-dilution problem using unofficial SEO files prompted Google's John Mueller to clarify that neither LLMs-author.txt nor Content-Signal headers carry any weight with search engines or AI crawlers.

The exchange highlights a growing trend of SEO practitioners reaching for technical solutions to problems that may require a fundamentally different approach — building a stronger digital presence rather than tweaking files that no major crawler actually reads.

---

What the Redditor Was Trying to Solve

The person behind the Reddit post faced a frustrating but increasingly common challenge. Their name is shared by two more prominent entities online, and that overlap dilutes the signals search engines and AI models use to identify them. As a result, their professional services become harder to surface when someone searches for them by name.

To counter this, the Redditor deployed two tools. First, they created a separate LLMs-author.txt file containing their job title, agency, location and area of practice written in plain sentences. Second, they added a Content-Signal header to their robots.txt file declaring ai-train=no, search=yes, ai-input=yes. They then asked the community whether anyone had run controlled tests on these Content-Signal headers to measure whether declaring intent at the header level changes anything.

The answer, according to Google, is a firm no.

Why This Problem Is Becoming More Common

Name-dilution is not an edge case. As more professionals build online presences, the likelihood of sharing a name — or even a brand identifier — with a more established entity increases. Search engines and AI systems that power modern search and discovery rely on weighted signals distributed across the web. When a more prominent entity dominates those signals, smaller or newer presences struggle to compete, regardless of their relevance to a specific query.

This is the root of what the Redditor experienced, and it is why the instinct to reach for a technical fix — however understandable — misses the point.

---

Neither Tool Is an Official Standard

Before Mueller even weighed in, the facts around both tools were already shaky. LLMs-author.txt does not appear to be based on any formal proposal or recognised standard. It is not an official specification, and no major search engine or AI crawler has confirmed support for it.

The Content-Signal header has a slightly more complicated origin. Cloudflare proposed a Content-Signal directive for robots.txt and later used identical syntax as an HTTP response header generated automatically through its Markdown for Agents feature. Markdown for Agents is a Cloudflare tool that serves a Markdown version of a webpage whenever a web client requests it. Despite that commercial implementation, the directive still has no backing from search engines or AI systems as an actionable signal.

Mueller responded directly to the Reddit post and was unambiguous.

"Google doesn't use llms.txt or llms-author.txt. I don't know of any other crawler or LLM confirming they're using these other than SEO tools."

On the Content-Signal directive, he was equally direct:

"It was made up by a CDN. As far as I know it has no effects whatsoever for any crawler or LLM. Using it just adds bloat and future maintenance to your robots.txt file."

Mueller noted that crawlers simply ignore directives they do not support, making the addition of these entries a maintenance burden with no measurable upside.

A Pattern Worth Recognising

This is not the first time unofficial or community-generated SEO specifications have gained traction before being dismissed by search engines. The SEO industry has a long history of practitioners adopting signals that sound credible but carry no confirmed weight. Understanding the core principles behind how search engines evaluate content and authority remains the most reliable foundation for avoiding these missteps.

Before adopting any new directive or file format, confirm it has formal support from at least one major search engine or AI crawler. Unofficial standards create maintenance overhead without delivering measurable benefit.

---

The Real Fix Has Nothing to Do With Technical Files

Mueller's response points to a deeper issue with how the Redditor framed their problem. They treated a visibility and identity challenge as a technical SEO problem when the actual solution lies elsewhere entirely.

Search engines and AI models build their understanding of who someone is by reading signals distributed across the web. When two or more prominent entities share a name, those entities tend to dominate simply because more of the web references them. No robots.txt directive or unofficial text file changes that underlying reality.

Building Authority Through Presence, Not Declarations

The path forward for the Redditor — and for anyone facing a similar identity challenge — involves expanding their footprint through the kinds of signals that search engines and AI systems already trust and act upon:

• Being interviewed on podcasts and YouTube channels
• Being mentioned in published articles and news coverage
• Contributing expert commentary to industry publications
• Appearing consistently in credible third-party sources

Each of these creates an external reference point that search engines can use to build a richer, more distinct understanding of who you are. It is the difference between telling a search engine who you are and showing it through a pattern of credible, consistent references across the open web.

Structured data helps too, but it works best when the broader web already contains consistent and credible references that reinforce it. The machine learning models that underpin modern search rankings are trained to weight external corroboration far more heavily than self-declared metadata — which is precisely why technical declarations in unofficial files carry so little influence.

Why This Approach Takes Longer — and Why That Matters

Building this kind of presence takes considerably longer than editing a robots.txt file, but it addresses the root cause rather than applying a patch that no crawler will ever read. The durability of the outcome is also categorically different. A robots.txt entry can be ignored the moment a crawler encounters an unsupported directive. A body of mentions, citations and appearances across reputable sources compounds over time and becomes increasingly difficult for any competing entity to displace.

Identity and entity disambiguation problems are fundamentally content and authority challenges. If your name or brand shares space with more prominent entities, the most durable solution is earning more mentions, citations and appearances across reputable sources — not relying on technical declarations that search engines are free to disregard entirely.

For further reading on how search engines and AI systems formally evaluate authorship and entity signals, Google's own Search Central documentation provides an authoritative reference point on what is and is not a supported signal.