Topic
AEO/GEO practice: managing brand citation, mention and inclusion inside AI-generated answers as a discovery surface distinct from ranked search results.
Semrush's seven-month index found that 36.7 to 45.3 of the UK's top 100 AI-mentioned brands also ranked among the top 100 cited domains. A dev.to summary says the US stayed under half too, but it quotes a US range of about 70 to 78.
Reality
Semrush found that among the 100 UK domains ChatGPT cited most, news and information sites drew 29.5% of citations, against 19.6% in the US. The index counts citations by site type, so a brand has to test whether press coverage moves its own rank.
Reality
Kingy AI Citation Index found the same provider's two runs overlapped 0.664 on cited domains for OpenAI, 0.515 for Claude and 0.319 for Gemini. One check cannot separate a lost citation from a different draw, so a change needs logged repeat runs before anyone calls it real.
Reality
LovedByAI's logs from 7 October 2026 attribute 703 of 1,022 ChatGPT-User 404s on 111 small-business sites to exploit probes. Counting ChatGPT's 404s by user-agent string mostly counts the scanner, so the client IP needs checking against OpenAI's published ranges.
Reality
Google is testing payments to dozens of invited publishers whose content meaningfully contributes to answers in AI Mode, AI Overviews and Gemini. Because the payout formula is unpublished, a citation count is only a proxy for what Google pays on.
Reality
Radyant's open dataset of 3,842 ChatGPT search runs found 28.6% of opening sub-queries named a vendor before any page was retrieved. Some AI search visibility is shaped by the model's category associations, ahead of any page it fetches.
Reality
FoxyInvoice prerenders every public Angular route to static HTML at build time so crawlers that skip JavaScript get the full page text. A build step and a one-line drift check take the place of the always-on SSR server its team declined to run.
Reality
The figure comes from an observational analysis the company itself says is not a random sample, and it points the money that moves an AI recommendation toward affiliate and earned media work sitting outside the SEO budget.
Reality
Andrew Higgins, who runs the AI search marketing firm Parsnipp, wrote that generative engine optimization is about two years old and that marketers cannot observe how people query the models. Ten experts, he said, give ten metrics.
Reality
The CLI does generative engine optimization, but GEO means geographic across most of the indexed web, so a model given only the name picked the wrong sense and then supplied a vendor to fit it. Closing the gap took three files.
Reality
PeakMetrics' AI Perceptions replays customer-chosen prompts through ChatGPT, Gemini, Claude, Grok and Perplexity and scores the answers against criteria the customer writes. Citation tracing works only when the model searched the web.
Reality
The number comes from a Forbes Tech Council post by a marketing agency director, and it is the only measurement in the record behind the claim that rank tracking and AI citation tracking now count different populations.
Reality
The Series D values the two-year-old marketing startup at $1.8 billion. Its prompt data comes from opted-in consumer panels, and the mention-frequency score a brand sees on the dashboard is a sample estimate.
Reality
Google restamped its spam policies on 15 May 2026 with a clause covering generative AI responses in Search. AI Overviews and AI Mode now fall under the same sanction sentence that has long covered ranking spam.
Reality
The operator of msgboard.dev logged three coordinated spam waves in three days, the last one carrying numbered steps that walk a reading agent through enrolling in a Lightning payment rail and holding a balance.
Reality
Sony Music and Warner have sued Anthropic over pirated music and lyrics, joining Universal. For publishers the live question is which crawlers still get the full text, and what evidence would show a training promise held.
Reality
A dev.to post sets out two ways assistant retrieval fails while ordinary search keeps working, one in robots.txt rules written per user-agent and one in copy that appears only after JavaScript runs. It offers no measurement.
Reality
Who absorbs the cost when an agent buys the wrong thing is still unsettled, and that leaves retailers deciding how far to integrate agent traffic before the price of an error is known. The funding is moving anyway.
Reality
The engines lean on Reddit, YouTube and LinkedIn, which means the pages deciding whether an assistant recommends your product sit on properties where your only interface is a post and somebody else's moderation queue.
Reality
Aggarwal et al. report visibility gains of up to roughly 40% inside generated answers, but the benchmark supplies the source documents, so the number scores selection and leaves crawl and index eligibility upstream of anything you can format.
Reality