ALT Cover: Being named is one number. Being named first is another. Brand Visibility and AI Visibility Score, explained.

ALT Two metrics: Brand Visibility is mentions divided by total responses times 100, for example named in 45 of 100 tracked answers is 45%. AI Visibility Score is weighted by where in the answer you are named.

ALT Bar chart of AI Visibility Score points per mention: named first 100, named second 50, named third 33.3. Not a percentage and not capped at 100.

ALT Read them together: high visibility with a low score means AI names you often but after others. Mention rate is calculated per model, then combined, so a strong model can hide a weak one.

ALT Cover: AI's most-cited health source is a research archive. The top domains in nearly 825,000 citations from US health answers, five AI engines.

ALT Bar chart of the most-cited domains in AI health answers, share of all 824,997 citations: PubMed Central 4.08%, YouTube 3.27%, google.com 3.16%, Reddit 1.35%, ScienceDirect 0.83%, Healthline 0.78%. Healthline is the highest-ranked medical publisher in the study.

ALT 33,669 citations for PubMed Central, more than five times Healthline's 6,456.

ALT Inside ChatGPT's answers, academic sources have an average position of 0.75, against 4.88 for Reddit and YouTube. Lower is higher up the answer.

ALT Cover: No sources in your ChatGPT chat? That's not the whole picture. Why some answers show no links, and how to check properly.

ALT Why an answer has no links: ChatGPT doesn't search the web for every answer and can answer simple or familiar questions from what it already knows. OpenAI doesn't publish a simple rule for which sources it picks.

ALT One chat is one answer: your chat shows what that one response cited. The same question with no citations in one model may trigger a web search in another.

ALT Check it properly: ask the same questions in several AI models, count mentions as well as citations, and repeat on a schedule.

ALT Cover: Google is putting more links inside its AI answers. What Google announced for AI Overviews and AI Mode on 15 October 2026, and what is only a test.

ALT Launching and rolling out: inline links, which Google says it has made continued and substantial increases to and will show higher up more often; news carousels rolling out to more locations and languages; related links at the end of some answers get image thumbnails.

ALT Still being tested: a dedicated section at the top of a response for the one or two official pages for a task, with why each link is useful, and clickable creator and publisher names on video results in AI Mode.

ALT What to do this week: mark 15 October on your AI visibility charts, check which of your pages each answer links to, and if you publish news, ask readers to add you as a Preferred Source.

ALT Cover: Google opened a fast lane for forum and social posts: the UGC Fresh Data Program, 8 October 2026.

ALT What it is: approved platforms send content and interaction signals straight to Google, separate from the Indexing API and crawling. No guarantee content appears in Search.

ALT Who can apply: platforms built on user-generated content with high volume, public posts tied to a creator profile, ideally sent within minutes, with structured data markup.

ALT It's for platforms, not brand sites. But the forums and communities that discuss your brand may qualify.

ALT Cover: Same 38 mentions: 15.2% or 31.7%. Share of voice and mention rate divide by different things.

ALT Mention rate is responses that mention you divided by all responses: 38 of 250 is 15.2%. Share of voice is your mentions divided by all mentions of you and your chosen competitors: 38 of 120 is 31.7%.

ALT One brand, one week, two numbers. Column chart of an illustrative worked example: mention rate 15.2% against share of voice 31.7%.

ALT Read them together: mention rate is absolute visibility, share of voice moves when you change the competitor list, citation rate counts responses citing a domain you own.

ALT Cover: Blocking AI training is not the same as blocking AI search. OpenAI, Anthropic and Google give training and search separate robots.txt controls; Perplexity's crawler is for search only.

ALT OpenAI: GPTBot crawls content that may be used for training. OAI-SearchBot surfaces sites in ChatGPT search. OpenAI says each setting is independent.

ALT Anthropic: ClaudeBot collects content that may be used for training. Claude-SearchBot indexes content for search. Claude-User fetches pages when a user asks a question.

ALT Perplexity and Google: PerplexityBot surfaces and links sites in Perplexity search and is not used to train foundation models. Google-Extended, a robots.txt token, covers Gemini training and grounding and does not affect inclusion or ranking in Google Search.

ALT Cover: No two AI engines read the press the same way. The outlet each one cites most, from 5.3 million AI citations.

ALT Bar chart of each engine's most-cited news outlet, as a share of its news-outlet citations: ChatGPT, TechRadar 18.5%; Gemini, PCMag 16.5%; Perplexity, PCMag 10.1%; Google AI Overviews, Mashable 9.4%; Google AI Mode, Mashable 8.7%. US questions, 12 April to 10 July 2026.

ALT Where each engine leans: ChatGPT is the most concentrated (TechRadar, Forbes, Tom's Guide); Gemini is a tech monoculture (PCMag, Mashable, TechRadar, Forbes); Google AI Mode and AI Overviews spread wider, both led by Mashable (AI Mode adds Yahoo and PCMag, AI Overviews adds PCMag and U.S. News).

ALT 2.8 times: Perplexity is where The New York Times over-indexes. It is Perplexity's 3rd most-cited outlet, with 5.7% of its news citations.