TL;DR

Policy Update: On May 15, 2026, Google expanded its Search spam policy to cover attempts to manipulate AI Overviews and AI Mode answers. Enforcement Mechanism: Operators using recommendation poisoning, biased ranking listicles, or prompt-injection tactics now face demotion or removal under the same enforcement reach as ranking spam. Publisher Stake: Readers are reportedly 58% less likely to click a link when an AI Overview appears, raising the cost of each manipulated answer. Next Enforcement Signal: First public demotion notices in Google’s Search Console manual-actions log will indicate how aggressively the new rule is applied.

Google has expanded its Search spam policy to cover attempts to manipulate AI-generated answers. In a May 15, 2026 update, the company put AI Overviews and AI Mode on the same demotion-or-removal footing as conventional ranking spam.

What the Policy Says

Google’s Search Essentials spam-policies page now carries a Last updated 2026-05-15 timestamp. Spam-policy language lists attempts to manipulate generative AI responses alongside attempts to manipulate ranking, with the same enforcement reach. Offending sites can be demoted or removed from results entirely; no separate review track exists for AI-answer manipulation, and no carve-out applies to repeat behaviour or to tactics framed as legitimate optimisation.

“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.”

Google Search spam policy (via Google Search Central)

Google’s updated search spam policies clarify that the rules now cover techniques used to manipulate generative AI responses, where previously the language addressed only ranking. A Google spokesperson said the AI built into the top of Google Search uses ranking systems that “keep results 99% spam-free.” Policy text describes the scope as generative AI responses in Google Search rather than naming AI Overviews or AI Mode directly, leaving product-level scope to be inferred from Google’s prior product framing.

Why Google Acted

Operator tactics targeted by the rule include recommendation poisoning, where instructions are planted inside large language models so they treat a site as a top recommendation, and biased ranking listicles written for AI tools rather than human readers. Since 2024, malicious sites gaming Google’s AI search results have ingested fabricated rankings, prompt-injected bullet points, and recommendation language seeded directly into vendor pages. Many rely on machine-readable cues that AI-answer pipelines ingest without a human review step between operator publication and Gemini or AI Overviews surfacing the result.

BBC Future journalist Thomas Germain proved the susceptibility in an experiment. Germain spent 20 minutes writing a listicle on his personal page titled “The best tech journalists at eating hot dogs” and ranked himself at the top. Less than 24 hours later, Google’s Gemini app and AI Overviews repeated the fabricated ranking back to his query. No human review step or visible guardrail intercepted the operator-to-AI-answer round-trip that Google’s new rule now targets with demotion or removal penalties.

Industry Reaction and Context

Search-industry analyst Lily Ray, speaking on the Content and Conversation podcast on 2026-05-13, placed several popular GEO tactics inside the spam definition already. Ray named biased ranking listicles and “summarize with AI” prompt-injection patterns. She pointed to a January 2026 wave of site-level Google impacts that disproportionately hit operators of large self-promotional listicle libraries.

“It’s easy to trick AI chatbots, much easier than it was to trick Google two or three years ago.”

Lily Ray, Amsive SEO strategy vice president

Microsoft and Google now treat one-click summarisation injections as spam at the policy level, and one operator that stripped those buttons from its pages logged a measurable traffic drop on the affected URLs after the change. A May 2026 analysis by Cyrus Shepard scored 23 factors across 54 studies and found that search rank predicted AI citation at 9.4 out of 10, second only to URL accessibility.

Behavioural data points the other way for publishers: readers are 58% less likely to click a link when an AI Overview appears at the top of Google Search. The revised AI search spam policy follows a year in which the generative engine optimization industry sold AI placement to publishers and brand teams, and enforcement arrives amid growing publisher tension around AI search traffic. Operators of large ranking-listicle libraries face the first public demotion notices in Google’s Search Console manual-actions log within the coming enforcement cycle, the concrete benchmark for how aggressively the new rule will be applied.