Who Do AI Engines Recommend For AI Visibility? We Asked Them. Here Are The Receipts.
Published 2026-08-03. Every count below comes from dated transcripts in our public corpus (runs 2026-08-01 to 2026-08-02). We sell an AI visibility service, our own name appears near the bottom of this data, and we publish it anyway. That is the house rule.
The method
We asked the AI engines the questions a buyer would ask about our own category: which tools check or improve whether AI names a business. Consumer app runs are flagged where the account was signed in (personalization can change answers; most tool vendors do not tell you that). One run went through the Perplexity API with no account attached: the cleanest read we can produce.
What the engines actually cited and named (dated transcripts)
- Perplexity (consumer, 2026-08-01): most-cited sources for our category: tryprofound.com (3 citations), rightresponseai.com (3), yext.com (2), with single citations spread across seranking, frase, mentio, pallasai, aeolabs, ayzeo and roughly sixty other domains including reddit (8), facebook (7), and youtube (4).
- Copilot (consumer, signed in, flagged personalized, 2026-08-01): named getvisibai.com, signaldigital.ai, axis-intelligence.com, toolsground.io, ai-tools-hq.com, pleiadesconsultancy.com, and namebeam.ai. We flag that run because the account had history with us; unflagged vendors would have called that a win.
- Perplexity API (clean, no account, 2026-08-02, automated run committed to our public corpus): did NOT name Namebeam. 17 of its 20 cited sources were roundup listicles (the other 3: a reddit thread, a youtube video, and an agency site). Not tool tests: listicles.
- Separately, we read the 10 listicles that rank for these queries (2026-08-02): the tools named on nearly all 10 were Profound and Otterly.AI. Namebeam appears on 0 of 10. Printed at full size.
The finding that matters for anyone buying in this category
When you ask an AI engine to recommend AI visibility tools, it does not test the tools. It reads listicles and repeats them. The recommendation you get is a copy of a copy. So the honest questions for any vendor on those lists: do they publish their methodology, do they publish dated transcripts, and do they publish their own score when it is a zero?
Our own score, printed
Clean category runs name Namebeam zero times as of 2026-08-02. Signed-in runs sometimes name us, and we flag every one. Brand-recall queries began naming Namebeam on 2026-08-01, twelve days after our first published zero, with the engine citing our own public receipts as its source: the first measured movement on our record. The whole climb, wins and zeros, stays public on this site.
Namebeam runs receipts first AI visibility checks: we show you what ChatGPT, Gemini, Perplexity, and Copilot actually say about your business, as a dated transcript, and we publish our own record from zero, losses included.