Skip to content
AI-Assisted Content5 min read

AI Is Routing Customers to Fake Brand Sites. New Data Shows 1 in 6 Fortune 500 Companies Are Affected.

The number came out of Sitecore Symposium in Orlando on November 12. Scrunch, a Sitecore company, ran a study across six AI platforms: ChatGPT, Perplexity, Gemini, Google AI Overviews, Google AI Mode, and Microsoft Copilot. Across all six, 17% of Fortune 500 companies had an impostor website cited in AI-generated answers about their brand.

Not just mentioned. Presented as the official source.

Three out of four times that a known impostor appeared in an AI answer, the AI described it as the brand's legitimate site. Scrunch estimates that more than one million AI answers per year contain a link to a website built to impersonate a real brand. That is not a narrow edge case. It is a failure in how AI assistants verify identity at scale.

Why This Is Different From Traditional SEO Brand Risk

Old search brand risk: a fake site shows up on page 8. A careful user might spot it.

AI brand risk: the fake site gets cited at the top of a conversational answer, delivered with the same confidence as any other researched fact. There is no position 8. No visible ranking signal. The AI says "here is the official website for Brand X" and links to a domain that has nothing to do with Brand X.

Companion research from ScrunchLabs adds another layer. More than 40% of AI sessions involve no web search beforehand and no click to the open web afterward. Compare that to traditional search sessions, where 21.5% are fully self-contained. When users stay inside an AI platform for the full interaction, they have no opportunity to cross-reference what the AI told them against a results page, a URL bar, or a second opinion. The impostor link is the only link they see.

What Peec AI Found the Day Before

One day before the Scrunch release, Peec AI published its Q3 2026 Primary Sources report, drawing on a Sapio Research survey of 700 marketing decision-makers across the US, UK, and Germany. The report found that 58% of marketers do not review which third-party sources AI platforms cite when users ask about their category. And 61% are not actively working to earn placement in those sources.

If your team does not know which third-party pages an AI cites when users query your brand, you also do not know if one of those pages is impersonating you.

The same report tracked a citation shift that happened with no announcement. On August 8, ChatGPT switched its default model to GPT-5.6. Within weeks, Reddit citations fell from a baseline index of 100 in Q2 to just 9 by mid-September. Product pages rose from 13.8% to 20.1% of all citations ChatGPT served. Content strategies built around third-party roundups and community platforms in Q2 were optimizing for a source mix that had already changed under them.

That volatility makes impostor tracking harder. A site that was not appearing last month may now be a primary citation by default.

What This Means for Content Strategy

A May 2026 Muck Rack study found that 84% of citations in ChatGPT, Claude, and Gemini came from earned sources, with paid content accounting for just 0.3%. AI platforms pull from what gets written about you, not from what you spend to place in front of users.

An impostor site exploits this exact mechanism. It is a piece of earned-looking content that talks about your brand, appears credible to a retrieval model, and shows up in citation results when users query your name or category. The AI has no way to call your legal team and ask if the domain is legitimate.

The best structural defense is a dense, cross-referenced citation footprint. When dozens of credible, independent sources consistently reference your brand's accurate information, the signal an AI retrieval system sees is harder to displace with a single fraudulent domain. That means press coverage, review site entries, industry news mentions, and consistent publishing on topics adjacent to your core offering.

This is not abstract. Sites that publish relevant industry content on a regular schedule build that footprint organically. Static sites with a single home page and no ongoing publication activity leave most of that citation space unoccupied and unmonitored.

What Scrunch Launched Alongside the Research

Scrunch released Citation Optimization at the same Symposium event. The tool identifies third-party pages cited in AI answers for each prompt a brand tracks, ranks them by influence and by how easily the brand can earn a place in them, shows where competitors appear instead, and drafts outreach using the brand's approved information.

A companion feature, Imposter Detection, automatically surfaces suspected impostor sites and provides the steps to initiate a takedown. Impersonation takedowns are free and unlimited across all plans. At launch, Scrunch integrated three distribution partners: Noble, Stacker, and Getfluence, to automate outreach and placement steps.

The minimum move for teams without a platform subscription is manual. Run your brand's core queries across ChatGPT, Gemini, Perplexity, and Copilot. Look at every URL the AI cites in its answer. Flag any domain that uses your brand name without being yours. Send the findings to legal.

The Broader Numbers

Peec AI's survey found that 89% of marketing teams increased AI search spending over the past year. Only 24% can demonstrate a return. The most common barrier cited by 33% of respondents: rapid platform change.

A channel that shifts citation patterns overnight after a model update and that may be routing some percentage of users to fake brand domains is genuinely difficult to measure. The teams Peec AI identified as most likely to prove ROI share one structural trait: they have integrated AI search into their broader marketing strategy rather than running it as a standalone pilot, and they have assigned at least one person to own it full-time. Companies with a dedicated full-time AI search role report trusted metrics at 57%, against 28% at companies where AI search is part of a broader role.

None of that changes the core finding from November 12. AI assistants are actively citing fraudulent sites as official brand sources for a significant share of the largest companies in the world. If that is true for the Fortune 500, the question for smaller brands is not whether the problem exists for them. It is whether anyone on their team has looked.


Sources: