For almost two years, a grey market quietly grew around one promise: pay us and we'll get your brand cited in AI Overviews, ChatGPT, and Perplexity. In May 2026, Google drew the line. That promise is now officially spam — and every SEO professional managing AI visibility work needs to audit their tactics against the new policy definition immediately.
This isn't speculation. On May 15, 2026, Google published a documentation update to Google Search Central that explicitly extended its spam policy definitions to cover AI-generated responses. The update was precise, deliberate, and aimed directly at a manipulation market that had been growing fast since AI Overviews became central to how millions of users experience Google Search.
I've managed SEO across healthcare, legal services, hospitality, and e-commerce for five years. I've watched the AI citation industry emerge, seen the pitches arriving in client inboxes, and tracked the tactics being sold as "GEO," "AEO," and "AI search optimisation" — some of it entirely legitimate, a worrying amount of it not. This article breaks down exactly what Google changed, why it matters now, and what you should audit this week.
The Exact Policy Change — Word for Word
The most important thing to understand: Google didn't create a separate AI spam rulebook. It expanded the definition of existing spam policy to cover AI surfaces. Here is the precise language change:
That single added clause — "or attempting to manipulate generative AI responses in Google Search" — is the entire change. And it's significant precisely because of what it doesn't do: it doesn't require new enforcement tools, new detection systems, or a new policy framework. Every mechanism Google already uses for link spam, scaled content abuse, cloaking, and site reputation abuse now applies to AI citation manipulation. The enforcement infrastructure was already in place. Google just pointed it at a new surface.
Why Google Moved Now — The Context Behind the Policy
This update didn't arrive without a backstory. Since AI Overviews became prominent in Google Search, an entire ecosystem of services emerged promising to get brands cited inside those AI-generated answers. The marketing language varied — "AI citation campaigns," "GEO optimisation," "AI visibility engineering" — but the underlying mechanics were often the same ones Google has penalised in traditional search for years: manufactured mentions, paid placements in third-party content, scaled publishing of low-value pages targeting AI retrieval, and in some cases explicit paid arrangements to influence AI summaries.
"I've watched the AI citation services market develop across the past eighteen months. Some of it is entirely legitimate — improving content structure, strengthening EEAT signals, making pages genuinely more citable. But I've also seen a growing number of pitches to clients that are, in everything but the terminology, the same link-building grey market repackaged for AI. The May 2026 policy update is Google doing what it did with paid links in 2012: drawing a clear, enforceable line before the manipulation becomes so widespread that it pollutes the AI experience the same way manipulated links polluted traditional search results."
What Is and Isn't Spam — The Critical Distinction
Google has been clear, and this is important: the policy does not target legitimate AI visibility work. It targets deceptive, manipulative tactics repackaged as AI optimisation. The distinction determines what you should continue doing versus what needs immediate review.
- 🔴 Paying for brand citations in AI Overviews or AI Mode
- 🔴 Scaled low-value pages targeting AI retrieval patterns
- 🔴 Separate AI-optimised page versions shown only to crawlers (cloaking)
- 🔴 Manufactured third-party mentions to create artificial citation authority
- 🔴 Near-duplicate pages targeting minor query variations to flood AI retrieval
- 🔴 Automated services generating self-referential "authority" content
- 🟢 Improving content quality and direct-answer structure for AI extractability
- 🟢 Strengthening real authorship signals and editorial credentials
- 🟢 Building genuine schema markup that accurately reflects page content
- 🟢 Earning real third-party citations through original research
- 🟢 Optimising for user intent so content genuinely merits citation
- 🟢 Measuring AI impressions via Search Console Generative AI reports
Which Existing Spam Policies Now Apply to AI Surfaces
<Because Google extended existing frameworks rather than creating new ones, every enforcement tool already in Google's arsenal now applies to AI manipulation. Here is the practical risk assessment across each major spam category:
| Spam Policy | How It Now Applies to AI Manipulation | Risk |
|---|---|---|
| Link Spam | Paying for placements in third-party content to drive AI Overview citations — same enforcement as paid link schemes | Critical |
| Scaled Content Abuse | Publishing high volumes of AI-generated or templated pages to flood AI retrieval with brand mentions | Critical |
| Cloaking | Showing different content to AI crawlers vs humans — flagged by Google and Bing as early as February 2026 | Critical |
| Site Reputation Abuse | Publishing third-party brand content on an authoritative host to borrow its trust signals for AI citation purposes | High |
| Doorway Pages | AI-targeted landing pages funnelling retrieval toward a brand without genuine user value | High |
| Keyword Stuffing | Overloading content with brand entity names primarily to trigger AI retrieval rather than serve user intent | Monitor |
Google's track record shows a gap between policy introduction and algorithmic enforcement at scale. Site reputation abuse was introduced in March 2024 but manual enforcement didn't begin until May 2024. This means the risk isn't zero today, but will grow as detection systems mature. The time to audit AI visibility workflows is now, before enforcement fully catches up with the policy.
Google's official guide to optimising for AI search — published alongside this spam policy update — contains the most clarifying statement the company has made on this entire topic: there is no separate AI index. The same content that ranks well in traditional search is the content that gets cited in AI answers. If your content isn't earning placement in the index for the relevant intent, no amount of AI-specific manipulation will get it cited in AI-generated responses.
Google also warned explicitly against creating separate content for every possible variation of how people might search, done primarily to manipulate AI responses. A high quantity of pages does not make a website higher quality — and it doesn't make it more citable in AI answers either. The logic is exactly what Google has always applied to traditional ranking: the reward changed from a ranking to a citation, and the rules followed it.
Your Audit Checklist — What to Review This Week
- 1Audit any AI citation services you're paying for. If the service involves paying for third-party brand mentions, paid inclusions in AI-curated content, or manufactured authority signals, document what it's doing and apply the test: would this be spam in traditional search? If yes, it's now spam in AI search too.
- 2Review your content production for scaled AI targeting. If your team is creating high volumes of near-duplicate pages targeting minor query variations specifically to appear in AI retrieval, that workflow is directly in scope for scaled content abuse policy enforcement.
- 3Check for AI-specific page versions. If you maintain separate markdown or structured versions of content designed to show different information to AI crawlers vs human visitors, that is now explicitly cloaking under the extended spam policy.
- 4Evaluate all third-party content placements. Guest posts, sponsored content, or editorial placements arranged primarily to generate AI citation authority — particularly on high-authority domains — are now in scope for site reputation abuse provisions.
- 5Redirect AI visibility budget to defensible content work. Original research, verified expert authorship, comprehensive topic coverage, real case studies and data — these create genuine AI citation authority without policy risk.
- 6Use Search Console's Generative AI reports as your measurement baseline. AI Overview impressions are now trackable separately from organic. Measure AI visibility improvements driven by content quality, not manipulation tactics that now carry explicit penalty risk.
Frequently Asked Questions
The Bottom Line
Google's AI citation spam update confirms what quality-focused SEOs have always known: AI search visibility operates by the same rules as traditional search visibility — not a new game with different standards, but the same game on a new surface. Audit your AI visibility tactics this week, redirect any manipulative spend toward defensible content work, and measure AI citation growth through Search Console rather than services that just drew a line in Google's own spam policy definition.
.webp)
Driven by advanced SEO expertise, deep marketing analytics, high-impact content strategy
With 5+ years of hands-on experience, I specialize in holistic search strategies that don’t just rank—they drive real, measurable business growth. I’ve worked across industries including healthcare, hospitality, legal, e-commerce, and professional services, helping brands dominate their target markets. My approach bridges the gap between raw data and creative execution. Every strategy I build is rooted in rigorous market analysis, structured SEO frameworks, and tailored content ecosystems—no templates, no shortcuts. Whether you’re a single-location brand or scaling across multiple cities, I create data-driven marketing systems designed to compound results and grow with you.
Concerned About AI Citation Spam Risks?
Get a complete AI citation audit, spam risk assessment, and actionable recommendations to improve your visibility in AI Overviews, AI Mode, and generative search results.
Request a Free AI Citation Audit →