Google warns publishers against ‘content chunking’ for AI-driven SEO

Google

• Google warns against rewriting content for AI systems instead of human readers
• SEO trend of "content chunking" does not improve Google Search rankings
• Google urges publishers to focus on clear, useful content for real users

Did our AI summary help?

Google has moved to shut down a growing SEO trend that encourages publishers to rewrite content for AI systems rather than human readers, calling the practice misguided and ineffective for search rankings.

According to a report by Ars Technica, senior members of Google’s search team have warned against a technique known as “content chunking,” which has gained popularity in the age of large language models. The strategy involves breaking articles into very short paragraphs, adding multiple subheadings—often phrased as questions—and structuring pages to resemble chatbot prompts rather than conventional articles.

The belief behind the trend is that AI systems such as Gemini prefer bite-sized information and are therefore more likely to ingest, summarise, or cite such content. Some SEO practitioners have also suggested that this approach could indirectly improve rankings on Google Search.

Google says that assumption is wrong.

On a recent episode of the company’s Search Off the Record podcast, John Mueller and Danny Sullivan addressed the issue directly. Sullivan said he has repeatedly seen advice urging creators to reformat their content specifically for large language models, but stressed that Google does not use such signals to determine search rankings.

“We don’t want you to do that,” Sullivan said during the discussion, adding that he had checked with Google engineers before making the statement. He explained that while AI systems may consume web content in different ways, Google Search continues to prioritise signals tied to real user behaviour—such as what people click on and how they engage with pages.

The podcast conversation, which begins around the 18-minute mark, highlights a broader concern inside Google about publishers chasing short-term SEO trends driven by fear of being left behind in an AI-dominated web. According to Sullivan, optimising content for machines rather than readers risks producing pages that feel unnatural and less useful, which can ultimately hurt visibility over time.

Google’s stance reinforces a long-standing message from the company: the most reliable way to perform well in search is still to create content for humans. Clear writing, logical structure, and genuine usefulness remain central, even as AI tools become more prominent across search and discovery.

For publishers and SEO professionals, the takeaway is straightforward—panic-driven optimisation for AI bots may be tempting, but Google says it’s not how search works, and not where creators should be focusing their efforts.