Google's AI Cites Crypto News Sites Four Times Less Than Rivals

Google's AI systems, including AI Overviews, AI Mode and Gemini, cite crypto news publishers far less often than independent AI platforms do. New research from growth marketing agency ICODA found Google's AI surfaces accounted for just 19.5% of 55,055 citations across 40 crypto news domains, while ChatGPT, Perplexity and Grok combined took 80.5%, a gap of 4.1 times, according to ICODA's research.

What changed?

ICODA measured 40 crypto news domains across six AI platforms and found Google's AI Overviews, AI Mode and Gemini citing the category 4.1 times less often than ChatGPT, Perplexity and Grok combined, according to the study. Six publishers received zero Google AI citations despite strong visibility elsewhere.

The research is one of the first to break out citation share by infrastructure owner rather than by platform brand, grouping everything Google runs against everything it does not. Across 55,055 total citations, Google infrastructure delivered 10,728, or 19.5% of the total, while independent infrastructure delivered 44,327, or 80.5%, the ICODA data shows.

Crucially, the research addresses the obvious objection that this could be a quirk of sample size. According to ICODA, the gap is not a sampling artifact: if the measurement tool were simply drawing from a smaller prompt set for Google's surfaces, every domain would hit the same ceiling. Instead, the pattern held consistently across all 40 domains, which is what makes the finding worth taking seriously rather than dismissing as noise.

Why does the timing matter?

This data lands in an already tense moment for the relationship between Google and the publishers it depends on for source material. ICODA's report notes that the findings arrive as scrutiny grows over how much traffic Google's AI features actually send publishers versus how much content they draw on without full attribution.

At the same time, publisher patience appears to be running thin more broadly, with several major outlets reportedly weighing whether to restrict access to Google's crawler over frustration with how AI features use their content, per ICODA's analysis. A study showing an entire content category gets systematically under-cited by Google's own AI adds fuel to that wider debate.

Why does this matter for your business?

If you run a website in a niche that depends on being named or linked by AI tools, this study is a reminder that "AI search" is not one system with one set of rules. Google's own AI products appear to lean on a narrower or more conservative set of sources for certain categories than independent players do, which means a brand thriving on ChatGPT or Perplexity could be almost invisible inside Google's AI Overviews and AI Mode, or vice versa.

Six publishers were cited nowhere in Google's AI, including Cointelegraph, cryptodaily.co.uk, Bitcoinist, NewsBTC and The Daily Hodl, according to ICODA's findings, despite strong presence on other platforms. These are established names in their sector, which suggests the issue is structural rather than a case of thin or low-quality content being filtered out.

For founders and marketing leaders, the practical takeaway is that you cannot treat "get cited by AI" as a single project. You need to check performance separately across Google's AI surfaces and the independent chatbots, because a strategy that wins on one can quietly fail on the other without anyone noticing until enquiries or revenue drop.

What should you do now?

Start by finding out where you currently stand on each platform, rather than assuming a good Google ranking translates into AI citations everywhere. Tools that track citation share by platform, and audits such as Sited's free AI visibility check at sited.online, can show you whether your brand is being picked up consistently or whether, like the crypto publishers in this study, you are strong on independent platforms but nearly absent from Google's own AI answers.

Second, treat Google's AI surfaces as a distinct optimisation target with their own quirks, rather than assuming SEO fundamentals automatically carry over. If your category shows a similar gap to the one ICODA found in crypto news, that is a signal worth investigating with your own data, not just a curiosity limited to one industry.

Third, keep an eye on the wider publisher-access story running alongside this one. If large publishers do start restricting Google's crawler over frustration with citation and traffic imbalances, the citation landscape could shift again quickly, and brands that rely heavily on being indexed and cited by Google's AI will feel that shift first.

Frequently asked questions

What is the ICODA study and who conducted it?

ICODA is a growth marketing agency for Web3 and iGaming brands. It published original research quantifying a gap between how often Google's AI systems cite crypto news publishers compared with independent AI platforms.

Does this citation gap only apply to crypto news sites?

The study specifically measured 40 crypto news domains, so the exact 4.1 times figure is category-specific. However, the underlying method of comparing Google's AI infrastructure against independent AI platforms could reveal similar or different gaps in other industries, which is worth checking for your own sector.

Why would Google's AI cite fewer sources in a category like crypto news?

The study does not give a definitive cause, only that the pattern held consistently across all 40 domains rather than appearing as random variation. Possible explanations range from stricter source-quality filtering to different indexing priorities, but ICODA's data only confirms the size and consistency of the gap, not the reason behind it.

Are publishers really considering restricting Google over this?

Separately from the ICODA study, reports have suggested that several major publishers are weighing whether to block Google's crawler entirely, reflecting broader frustration with how Google's AI features use their content. This context adds weight to ICODA's citation findings rather than being part of the study itself.

What should a small brand do differently based on this?

Do not assume that being cited well by ChatGPT or Perplexity means you will also show up in Google's AI Overviews or AI Mode, and check each platform separately. If you find a similar gap to the one seen in crypto news, focus extra effort on the structured, authoritative content signals that tend to earn citations specifically within Google's AI surfaces.