Eight AI visibility myths to stop believing

The myths

1. 1Myth: adding llms.txt gets you cited. Reality: it is a proposal with uneven support, and a helpful extra at most.
2. 2Myth: schema markup guarantees inclusion. Reality: it helps machines understand a page, and Google does not promise inclusion.
3. 3Myth: there is a secret ranking formula. Reality: platforms do not publish one, and answers are sampled.
4. 4Myth: one screenshot proves your position. Reality: one answer is one draw.
5. 5Myth: a single score tells the whole story. Reality: rates differ by platform, topic and location.
6. 6Myth: you can pay for a recommendation. Reality: undisclosed paid placement breaches rules and trust.
7. 7Myth: AI visibility is separate from SEO. Reality: they share foundations.
8. 8Myth: once fixed, always fixed. Reality: answers change as models, sources and your own facts change.

Myth: you need special files or markup to appear in AI Overviews

Google's own documentation addresses this directly:

You don't need to create new machine readable files, AI text files, or markup to appear in these features.

This does not make structured data or an llms.txt file pointless. They can help other systems understand a site, and structured data helps Google understand a page. It means no special file is a requirement for Google's AI features, and adding one is not a shortcut to being cited.

Myth: blocking AI crawlers protects you from AI answers

Crawlers have different purposes, and the major providers document them separately. OpenAI states that each setting is independent, so a site can allow OAI-SearchBot to appear in ChatGPT search while disallowing GPTBot, which relates to training. Perplexity states that PerplexityBot is not used to crawl content for foundation models. Blocking everything at once can therefore remove a site from search-connected answers without doing what the owner intended.

Myth: robots.txt controls every fetch

Two providers say plainly that it does not. OpenAI notes that because ChatGPT-User actions are initiated by a person, robots.txt rules may not apply. Perplexity says Perplexity-User generally ignores robots.txt because a user requested the fetch. If content must be private, protect it with authentication, not with a robots.txt line.

Myth: one dominant assistant means one place to optimise

ChatGPT has the largest share of generative AI website traffic, but the mix is shifting. Similarweb's August 2026 figures put ChatGPT at 55.5% of tracked chatbot website traffic, against 73.3% a year earlier, with Gemini at 25.6% and Claude at 9.3%. Website traffic leaves out apps and system-level entry points, so it is a partial view. It still shows a market that is moving, and a brand should measure the assistants its own customers use.

Myth: AI has ended the value of clicks

The Pew Research Center found that Google users clicked a traditional result link in 8% of visits when an AI summary appeared, against 15% when one did not. That is a real drop in clicks on that page. It is not evidence that clicks or visibility have stopped mattering, and Pew's data cannot tell us how many people acted on a brand name they read in the summary.

These figures come from Pew Research Center's analysis of the browsing data of 900 U.S. adults during March 2025. They describe Google searches in one country and one month, and they measure clicks, not influence. A person who reads a brand name in an AI summary and searches for it later will not appear in these numbers at all.

The pattern behind the myths

Most myths share a shape: a real observation is stretched into a rule. A rate falls, a tactic seems to help once, a vendor asserts a formula. The remedy is the same each time. Ask what was measured, how many observations sit behind it, and whether the claim would survive being repeated.