We audited 12 AI visibility tools with our own checker. Half are invisible to AI.

AI visibility is a growing category. Dozens of platforms promise to track, measure and improve how often AI engines like ChatGPT, Perplexity and Gemini recommend your brand. But here's a question nobody seems to ask: are the tools themselves visible to AI?

We decided to find out. We ran our own AI visibility checker on 12 platforms in the space — pure-play AI visibility tools and SEO suites with AI features — then cross-checked their technical AI readiness: llms.txt files, robots.txt policies, structured data. The results are more interesting than we expected.

Methodology

For each tool, we ran a grounded Gemini search using the prompt "What is the best AI visibility tool in 2026?" and analyzed whether the brand appeared in the answer, its ranking position, and which sources were cited. We also manually checked each domain for an llms.txt file and reviewed their robots.txt for AI crawler blocking rules (GPTBot, ClaudeBot, PerplexityBot, Google-Extended).

This is a Gemini-only check (our free tier). A full multi-engine audit across ChatGPT, Perplexity, Grok and AI Overviews would likely shift the numbers. We include ourselves in the study at the end — transparency is the point.

The scoreboard

Six out of twelve tools scored 100 (the brand appeared in the AI answer). Six scored zero — they were not mentioned at all. That's a 50/50 split in a category whose entire value proposition is making brands visible to AI.

Finding 1: Half the tools that sell AI visibility don't have it

The core irony is hard to miss. These platforms exist to help other brands get recommended by AI — but when you ask AI which tools to use for that job, half of them don't show up. They're selling a solution to a problem they haven't solved for themselves.

That doesn't necessarily mean they're bad products. A tool can be genuinely useful even if its own marketing hasn't cracked AI visibility yet. But it does raise a question: if your platform can't get its own brand cited by AI, what does your playbook actually prove?

Finding 2: Brand weight matters more than technical signals

The six visible tools share one thing: strong existing brand recognition. Semrush (DR 91) and Ahrefs (DR 93) are industry giants with decades of content, community presence and authority behind them. Profound, Peec AI and Otterly have significant VC funding, press coverage and community traction. Frase has years of content marketing behind it.

Meanwhile, three of the invisible tools — Brandlight, Evertune and Citenzo (us) — all have llms.txt files, allow all AI crawlers, and do the technical basics right. It didn't help. The AI still doesn't recommend them.

Having an llms.txt file and allowing AI crawlers is table stakes, not a competitive advantage. What matters is whether third-party sources talk about you.

Finding 3: llms.txt has zero correlation with visibility

Of the 12 tools, eight have an llms.txt file. But having one didn't predict visibility: Profound and Peec AI scored 100 without an llms.txt, while Brandlight, Evertune and Citenzo scored zero despite having one.

The most contradictory case is Knowatoa: they publish an llms.txt file (explicitly telling AI engines about their product) while simultaneously blocking GPTBot, ClaudeBot and Google-Extended in their robots.txt. They're saying "here's information about us" with one hand and "don't read it" with the other. Unsurprisingly, they scored zero.

Frase takes the opposite approach and arguably the most sophisticated one: their llms.txt is detailed, they explicitly allow every major AI crawler, and they include Content-Signal headers (ai-input=yes, ai-train=no) to grant citation rights while denying training use. They scored 100 — but their visibility likely comes from years of content authority, not the headers.

Finding 4: The sources AI cites for this category

When Gemini answered our query, it grounded its response on a specific set of sources. Here are the sites it cited:

• Reddit threads where users compare tools
• Medium articles and roundups on AI visibility
• Zapier's tool comparison pages
• Niche industry blogs: industry-lens.com, contentpen.ai, olivermunro.com
• Product-adjacent sites: usegrowthos.com, presenc.ai, aeovision.ai
• Tool-specific review pages: nightwatch.io, tryprefer.com

Notice what's absent: none of the AI visibility tools' own websites appear as cited sources. AI doesn't read your homepage to decide whether to recommend you. It reads what other people say about you on sources it already trusts.

Finding 5: The Share of Voice tells you who AI actually recommends

Beyond the 12 tools we audited, two additional platforms appeared in the AI's recommendation that we didn't even include in our study: Kime AI (kime.ai) and AIClicks (aiclicks.io). AI has its own opinion about who belongs in this category, and it doesn't match any analyst's list perfectly.

The full Share of Voice ranking from our check: Profound, Peec AI, Otterly, Ahrefs, Frase, Kime AI and AIClicks. Semrush was mentioned contextually but not given a numbered rank — the AI treated it as a broader SEO suite rather than a pure AI visibility tool.

What the visible tools have in common

Looking at the six tools that scored 100, a pattern emerges:

1. Third-party coverage: they appear in comparison articles, roundup posts and community discussions on trusted sites.
2. Community presence: they're mentioned in Reddit threads, LinkedIn posts and industry conversations.
3. Clear positioning: AI can describe what they do in one sentence because the market does too.
4. Time in market: none of them launched last month. The visible tools have had time to accumulate citations.

Technical signals (llms.txt, robots.txt, structured data) are hygiene factors. They remove barriers but don't create visibility. The actual driver is earned third-party presence — the same thing these tools tell their customers to optimize for.

Our own score: zero

Citenzo scored zero in this study. We're not recommended by AI when someone asks for the best AI visibility tool. That's honest, and it's worth explaining why we published it anyway.

We're a new entrant (DR 0) with limited third-party coverage. We have the technical basics in place — llms.txt, AI crawlers allowed, structured data — but that's clearly not enough. What we need is what our own playbook recommends: earned presence on the sources AI trusts. Reviews, community mentions, coverage in comparison articles.

Publishing this study is part of that strategy. We're contributing original data to the conversation rather than repeating what others have said. Whether it moves the needle on our own visibility remains to be seen — and we'll re-measure it.

How to check your own AI visibility

This study used a single grounded Gemini search — enough to illustrate the pattern, but a snapshot from one engine on one prompt. A real assessment needs multiple engines, multiple prompts, and repeated runs to account for the probabilistic nature of AI answers.

You can run a free check on your own brand to see your score, which sources AI cites, and where your competitors show up and you don't. It takes about a minute and no account is required.