On May 15, 2026, Google updated its spam policies to explicitly cover attempts to manipulate AI Overviews and AI Mode. Here is what every service business needs to do now.
Ido Cohen · Published 2026-05-17 · SEO & Search
Google officially declared AI search manipulation a spam violation on May 15, 2026 — and if you've been using any of the trendy "GEO hacks" floating around, your rankings and AI visibility are now directly at risk. This isn't a vague algorithm tweak. It's a policy change that puts generative engine optimization (GEO) tactics on the same enforcement track as link spam and keyword stuffing, with penalties ranging from rank demotion to outright removal from Search. If you own a plumbing company, a dental practice, a law firm, or any other service business trying to show up in AI Overviews and AI Mode, this update changes your playbook.
Google didn't launch a new algorithm on May 15. It did something arguably more significant: it rewrote the foundational definition of spam to cover AI.
According to Search Engine Land, which first spotted the change, Google's spam policies page — last updated May 15, 2026 — now includes a new clause. The updated introductory line states, in full: "In the context of Google Search, spam refers to techniques used to deceive users or manipulate our Search systems into featuring content prominently, such as attempting to manipulate Search systems into ranking content highly or attempting to manipulate generative AI responses in Google Search."
That bolded phrase is new. Previously, the definition only referenced manipulating "Search systems into ranking content highly." The addition of generative AI responses explicitly brings AI Overviews and AI Mode — the two surfaces where most service-business visibility is shifting — under the same spam umbrella as every classic black-hat SEO tactic Google has historically cracked down on.
The Verge reported that Google may penalize sites using tactics like biased listicles or "recommendation poisoning" with lower rankings or removal from results entirely. Google's own policy page confirms: violations can trigger rank demotion or removal, detected through both automated systems and human review.
This is not a theoretical risk. It is policy.
GEO stands for Generative Engine Optimization — a term that emerged over the past 12 months as marketers tried to reverse-engineer how to appear inside ChatGPT, Perplexity, Google's AI Overviews, and AI Mode.
The legitimate version of GEO is straightforward: create authoritative, well-structured content that AI systems naturally want to cite. That's just good SEO repackaged for a new interface.
The illegitimate version — the kind Google just outlawed — looks like this:
Gizmodo's coverage confirmed Google's policy explicitly enumerates AI-mass-produced low-value pages, cloaking content, and expired domain abuse as examples. Search Engine Land noted that there is a lot of advice out there about appearing in AI search engines, and that some of that advice may now directly violate Google's spam policies.
For service business owners, the practical translation is this: if an agency or consultant sold you a "GEO package" that involved any of the above, you are now exposed.
Service businesses have more at stake in AI search than almost any other category. Here's why this matters specifically for plumbers, dentists, lawyers, HVAC contractors, med spas, financial advisors, and real estate agents:
AI Overviews dominate high-intent queries. When someone types "emergency plumber near me" or "best family dentist in Austin," AI-generated answers now appear at the top of the results page — often above the local 3-pack. Being cited there is worth real leads.
The stakes of getting delisted are massive. Unlike an e-commerce site that can survive on paid ads if organic traffic drops, most service businesses get 60-80% of their inbound leads from local and organic search. A manual action or rank demotion from Google is a direct hit to the phone ringing.
The GEO advice ecosystem is polluted. SOCi's 2026 Local Visibility Index found that AI platforms are far more selective than traditional local search — only 1.2% of locations were recommended by ChatGPT, 11% by Gemini, and 7.4% by Perplexity, compared to 35.9% showing in Google's local 3-pack. That data created a desperate market for "shortcuts" to AI visibility. Agencies rushed to fill it with tactics that are now officially spam.
Penalties compound. A manual action from Google's human review team doesn't just hurt your AI Overview appearances — it can suppress your entire domain across traditional organic results, Maps, and paid placements simultaneously.
Here's a plain-English breakdown of where the line falls:
The dividing line is intent. If a tactic is designed to deceive the AI rather than genuinely serve a human customer, it's now spam.
Here's the good news for service businesses who've been doing things right: nothing about the legitimate path to AI visibility changed on May 15. What changed is that the illegitimate shortcuts just got closed off.
According to Google's own AI search documentation, AI Overviews and AI Mode pull from content that demonstrates strong E-E-A-T signals — Experience, Expertise, Authoritativeness, and Trustworthiness. For a local service business, that means:
Real credentials, prominently displayed. A licensed HVAC technician with 15 years of experience in Dallas should have that information clearly on every relevant page — not buried in a footer, but in the body content where AI can parse and credit it.
First-person specificity. AI systems increasingly favor firsthand accounts over generic summaries. A plumber who explains exactly how they diagnose a slab leak in older Houston construction is more citable than a generic "what is a slab leak" explainer.
Consistent business data across the entire web. Business profile information was found to be only about 68% accurate on ChatGPT and Perplexity in SOCi's 2026 index, compared to 100% accuracy on Gemini (which is grounded in Google Maps). That gap creates citation errors. Accurate, consistent NAP data (Name, Address, Phone) across your Google Business Profile, your website, and every directory is the single highest-ROI move for AI visibility.
Genuine community signals. Google's recent AI Mode updates already showed that the platform is pulling in Reddit discussions, forum posts, and community content. When real customers mention your business positively in local Facebook groups or neighborhood forums, that organic social proof now feeds directly into AI-generated recommendations. You cannot fake this — but you can earn it by delivering great service and asking satisfied customers to talk about you online.
You have a narrow window to audit your current situation before Google's automated detection systems flag anything. Here are five specific actions to take before Friday:
1. Audit your existing content for mass-generated pages. If your agency produced more than 30 location-variation pages using AI tools in the past 12 months, pull a sample of 10 and read them honestly. Do they say anything unique and useful about that city or zip code, or are they 90% boilerplate? Any page that's purely a keyword vehicle with no human value should be consolidated or rewritten now — not after a manual action.
2. Cancel any GEO "package" you don't fully understand. If a vendor can't explain in plain English exactly what they're doing to improve your AI visibility — and what content they're producing — stop the engagement and get a refund conversation started. "We optimize your content for AI citation signals" is not a sufficient explanation. Ask for the specific deliverables and review them against Google's spam policy list.
3. Strengthen your Google Business Profile. This is the highest-accuracy data source feeding Gemini, Google's AI Mode, and AI Overviews. Update your services list with specific descriptions (not just "Plumbing Services" but "emergency leak repair, water heater installation, drain cleaning"). Add photos taken this month. Respond to your three most recent reviews today.
4. Create one genuinely expert piece of content this week. Pick the question your customers ask most that no competitor in your market has answered in depth. Write a 600-900 word answer in your actual voice, with specific details that only someone with your expertise would know. Publish it. That single piece of real content is worth more than 50 AI-spun pages.
5. Search your own business name in AI Mode. If you have access to Google's AI Mode (available to users in the U.S. via Search Labs), run five queries a customer would use to find your type of service in your city. Screenshot what comes up. If your business doesn't appear, you now have a clear benchmark. If it does appear but with inaccurate information, you have an urgent accuracy problem to fix — and that fix starts with your GBP and website, not any shortcut.
The businesses that will win AI search over the next 18 months are the ones who recognized early that AI systems are getting better at understanding genuine expertise versus manufactured signals. May 15 was Google making that reality official policy.
---
What exactly did Google change in its spam policy on May 15, 2026?
Google updated the opening definition of spam in its Search spam policies to explicitly include "attempting to manipulate generative AI responses in Google Search." Before this update, the definition only covered manipulating traditional ranking systems. The change means that tactics designed to force appearances in AI Overviews and AI Mode are now treated as spam violations, subject to the same penalties as link spam, cloaking, and keyword stuffing.
What is GEO and is it now completely banned?
GEO stands for Generative Engine Optimization — the practice of optimizing content to appear in AI-generated search answers. The legitimate version (creating genuinely authoritative, well-structured content that AI naturally cites) is not banned and remains effective. What Google has outlawed is manipulative GEO: hidden AI-targeting text, fake recommendation pages, mass-generated low-value location content, and other deceptive tactics designed to game AI responses rather than serve real users.
What penalties can my service business face if Google flags my content as AI manipulation spam?
Google's spam policy page states that sites violating its policies "may rank lower in results or not appear in results at all." Violations are detected both by automated systems and human reviewers. A manual action — the most serious outcome — can suppress your entire domain across organic search, Google Maps, and AI features simultaneously. For a service business that depends on local search for leads, this is an existential threat to inbound volume.
How do I know if an agency I hired is using tactics that violate this policy?
Ask your agency for a list of every piece of content and every technical action they've taken to improve your AI visibility in the past six months. Review any location-variation pages they created: if they were bulk-generated with AI and contain no unique local expertise, they're at risk. Also look for any technical SEO changes involving hidden text, structured data manipulation beyond standard schema, or link-building specifically framed as "AI authority building." If they can't provide clear documentation, treat that as a red flag.
Does this change affect my Google Ads campaigns or just organic results?
This specific spam policy update applies to organic search results, AI Overviews, and AI Mode — not to paid Google Ads placements. However, Google Ads has its own quality and policy enforcement systems, and a domain that receives spam penalties in organic search can sometimes see quality score degradation in paid campaigns as well. The safest position is to build genuine authority across both channels rather than relying on any single paid or organic tactic.
---
Sources: