What are the signs your brand is invisible in AI search engines?

AI invisibility shows up as a handful of observable symptoms. Each maps to one of four stages: what the model learned in training (parametric), whether crawlers can fetch your pages (access), whether your page is pulled in for a query (retrieval), and whether the model uses and cites it (generation). You control access fully, retrieval partly, and the other two not at all.

The fastest way to see the first sign is to ask the questions a buyer would ask. The free checker on this site runs up to three buyer questions and shows which engines name you. It is a snapshot, so treat it as a starting point.

Each engine draws on its own index and its own crawler. OpenAI's crawler documentation says OAI-SearchBot controls ChatGPT search and GPTBot controls training only, and that "each setting is independent of the others." A site that blocks OAI-SearchBot will not be shown in ChatGPT search answers. Blocking GPTBot does not remove you from ChatGPT search, and allowing it does not get you in.

Google states that a page must be indexed and eligible to show with a snippet to appear as a supporting link in AI Overviews or AI Mode, with no additional requirements. So a Google index problem shows up in Google's AI features and nowhere else.

Otterly's 2,000 AI blogs experiment shows how far engines can diverge. After Google deindexed both test sites, Microsoft Copilot, which grounds on Bing, produced 2,718 of the 3,934 citations Otterly tracked for one site. ChatGPT cited the sites effectively zero times, even while they ranked in Bing.

If one engine cites you and another ignores you, read why Perplexity can cite a site that ChatGPT skips.

The causes behind access signs are documented by the operators themselves, and fixing them removes a block.

Crawler 403s and challenge pages

OpenAI's help pages say eligibility needs both a robots.txt allow for OAI-SearchBot and a host or CDN that lets OpenAI's published IP addresses through. A firewall can return 403 to a bot that robots.txt allows, and nothing in your dashboards will flag it. From 2026-09-15, Cloudflare blocks Training and Agent crawler categories by default on ad-displaying pages for new domains and existing free-tier customers, so access can change without anyone editing your site.

Facts that only exist after JavaScript

Vercel's crawler study found that ChatGPT and Claude crawlers fetch JavaScript files but do not execute them, so they cannot read client-side rendered content. Googlebot renders JavaScript, so this sign appears in ChatGPT, Claude and Perplexity answers but not in Google's.

Suppression tags

Google documents noindex, nosnippet, max-snippet and data-nosnippet as controls that keep content out of its AI features. Bing says content marked data-nosnippet stays indexed but is excluded from snippets and AI summaries. These tags often arrive by accident through a template.

The access fixes compared with content rewrites page covers why these levers outrank on-page tactics.

Some symptoms look like invisibility but measure something else.

• No AI referrals. OpenAI says ChatGPT adds utm_source=chatgpt.com to referral links, but other engines tag inconsistently and Google AI Overviews send no referral at all. Low referrals can mean poor tagging, not absence.
• Answers with search turned off. When an engine answers from memory, only its training data counts. Content published after the model's cutoff cannot appear, and nothing you change this quarter reaches that answer.
• A blended score. One number averaged across engines and retrieval states can hide a total failure on one engine. That is one reason to be wary of share of voice packages.
• Losing to big brands. Research on AI citations finds a strong association with brand prominence. Being outranked by household names is a different problem from being unreachable.
• Third-party citations. When engines cite review sites or marketplaces instead, being named on those pages is the route in. See why AI links to resellers instead of your site and how third-party mentions shape AI answers.

Work from the gates you control toward the ones you don't.

1. robots.txt. Confirm OAI-SearchBot, PerplexityBot, Claude-SearchBot and Googlebot are allowed. A blanket AI block is the common error.
2. Firewall and CDN. Request a page as each AI crawler and record the status code. The AI Crawler Checker shows which crawlers your firewall turns away.
3. Rendering. Compare the raw HTML with the rendered page. Your prices, product names and key facts should appear before JavaScript runs.
4. Suppression tags. Search templates for noindex, nosnippet, max-snippet:0 and data-nosnippet.
5. Index status per engine. Check Google Search Console and Bing Webmaster Tools separately. One being healthy says nothing about the other.
6. Facts. State each key fact once, plainly, in fetchable HTML, and make sure your pages don't contradict each other.

The full AI SEO checklist expands each step. None of these checks guarantees a citation. They remove the blocks that make one impossible.

Will adding schema markup fix AI invisibility? The evidence does not support it. Google says structured data isn't required for generative AI search. Ahrefs' matched test of 1,885 pages found AI Overview citations fell 4.6% after schema was added, though those pages were already heavily cited. Existing markup does no harm, but it is not a lever for AI answers. Check access, rendering and indexing first.

Does blocking GPTBot hide my brand from ChatGPT search? No. OpenAI documents GPTBot as the training crawler and OAI-SearchBot as the crawler for ChatGPT search, and says each setting is independent. You can block GPTBot to opt out of training and still appear in ChatGPT search, as long as OAI-SearchBot is allowed and your firewall lets it through.

Why does ChatGPT describe my brand with old information? Often the answer came from training data rather than a live search, and that data has a cutoff date. As of October 2026, ChatGPT's models carry different cutoffs depending on which model handles the request. If ChatGPT did search, check whether your current facts appear in raw HTML, since its crawlers do not run JavaScript.

Can I see my position inside Google's AI answers? No. Google's Search Console report for AI features shows impressions only, with no clicks, click-through rate or position. Any tool reporting a rank inside AI Overviews has no Google data to check it against. Presence, absence and citation counts per engine are the honest measures.

Is hiding text for AI crawlers a way to get noticed? No. Hidden text, cloaking and prompt injection aimed at AI crawlers are illegitimate. Research shows manipulated content can sway some AI search agents once retrieved, but engines treat it as abuse and it puts your brand at risk. Fix access and state your facts plainly instead.