Google updates anti-spam rules for AI in Search

Google's new definition of spam

On May 15, 2026, Google published a revision of its spam policy. The update was reported by The Verge and picked up by Search Engine Land. The key change is that the concept of spam now extends to the artificial intelligence systems built into search.

The new wording explicitly states that spam includes attempts to manipulate Google Search's generative systems, such as AI Overview and AI Mode. So it's no longer enough to avoid traditional practices like keyword stuffing or link farming. You now also need to consider the impact of your content on the LLMs that compose automated answers.

This shift is significant. Until now, the spam policy focused mainly on protecting the human user. With this update, Google also protects the integrity of its own AI system.

What manipulating AI answers means

Two main categories of techniques are now explicitly banned. The first is producing "best-of" listicles built to influence AI-generated recommendations. The second is recommendation poisoning, meaning the injection of distorted content into the text corpora that LLMs use as a source.

Google also flags attempts to embed hidden instructions in text, a practice similar to prompt injection, to steer AI Overview answers toward certain products or brands. According to Google Search's official documentation, these techniques will be handled with the same measures applied to traditional spam.

In particular, affected sites risk being demoted in organic results and excluded from AI-generated answers. The potential impact is therefore double that of a classic penalty.

Why Italian SMEs are exposed to this risk

Italian small and medium-sized businesses often operate in competitive niches where differentiation comes through comparison content, buying guides and product reviews. These are exactly the content types most at risk under the new policy.

But the problem doesn't only affect those who knowingly used improper practices. Many SMEs rely on agencies or freelancers who produce content optimized with logic that is now obsolete. As a result, even pages created in good faith can fall into the categories Google has flagged.

According to McKinsey's analyses, the share of searches handled by AI systems in Search is set to grow significantly by 2027-2028. The risk of being excluded from AI answers will therefore become increasingly relevant to overall organic visibility.

For SMEs that have invested in SEO strategies in recent years, this update calls for a structured review of existing content. It's not a minor task.

The operational impact on your content strategy

The new policy changes the boundaries of what counts as quality content. In the past, a well-structured comparison page, with data, tables and optimized anchor text, was considered a solid SEO asset. Today, if that same page is built to artificially steer AI answers, it becomes a liability.

Google also hasn't provided an exhaustive list of banned patterns, which leaves some room for interpretation. The most prudent strategy is therefore to produce content that answers a real user question, with verifiable sources and no rhetorical forcing.

In practical terms, these are the areas to monitor first:

• "Best products" or "top 10" pages built without transparent editorial criteria
• FAQ sections optimized with answers designed to appear in AI Overview
• Content that systematically replicates the queries tied to generative answers
• Texts with repetitive structures meant to saturate the context of LLMs

The SEO copywriting team at SHM Studio is already updating its editorial frameworks to reflect these criteria.

What to do now: a phased approach

First, you need to audit your existing content. The goal is to identify pages that could fall into the categories flagged by the new policy. Then, set intervention priorities based on traffic and the strategic importance of each page.

You should also review your editorial briefs, both internal ones and those you give to external suppliers. Content production must focus on real usefulness for the user, not on saturating AI-related queries. This principle isn't new, but the new policy makes it binding in a more explicit way.

Finally, we recommend checking Google Search Console more often over the coming weeks. Signs of manual or algorithmic penalties may surface a few weeks after the policy update date.

If you run integrated digital marketing campaigns, you should also check that your organic and paid content are consistent. The landing pages of your Google Ads campaigns are subject to the same quality assessments.

The open questions: what Google hasn't clarified yet

The updated policy leaves some important questions open. In particular, it's unclear how Google tells content genuinely optimized for AI apart from content built to manipulate it. The line is thin and, for now, left to algorithmic interpretation.

Likewise, it isn't specified whether penalties will affect only presence in AI Overview or traditional organic ranking too. This detail is crucial for anyone who has built their visibility on both channels.

At SHM Studio, we're following the evolution of this policy closely. The guidance we give our clients is updated in real time, combining the analyses of our specialists in AI applied to marketing with Search Console data and Google's official communications.

If you want to dig deeper into content quality in the age of AI, the SHM Studio blog regularly publishes updates in the resources and insights section.

Outlook for 2027-2028

The May 2026 update is probably just the first of many. As AI becomes more deeply integrated into search, Google will increasingly need an internal regulatory framework to deal with attempts to manipulate generative models.

SMEs that invest today in editorially solid, verifiable content will therefore gain a competitive edge in the coming years. By contrast, those who keep producing content optimized only for traditional metrics risk losing visibility in both organic ranking and AI answers.

SEO and content marketing strategies therefore need to evolve toward a more rigorous editorial model. This also applies to content distributed on channels like LinkedIn, where consistency between your organic presence and the content Google indexes matters more and more.

If you want to review your strategy, you can request an initial consultation through the SHM Studio contact page. Our team is available for a preliminary analysis with no obligation.