Google Tests AI Headlines & Spam Update

This week’s developments in search highlight significant shifts in how content is discovered and displayed, with Google testing new headline formats and completing a rapid spam update. The search giant is also introducing new structured data options for labeling automated content, while Bing provides publishers with deeper insights into AI-driven citations.

Google has confirmed a limited test of AI-generated headline rewrites in its traditional search results. The company described the experiment as small and narrow, using language similar to that employed before it reclassified a similar feature in Google Discover. Notably, these rewritten headlines do not include any disclosure that Google has altered the original title. While Google stated that a broader launch might not use generative AI, it did not clarify what alternative technology might be used.

This move represents a meaningful evolution in how Google presents content. Previous headline adjustments primarily focused on matching query intent or fixing technical issues like truncation. The current test appears to use AI to rewrite for engagement, with documented examples showing changes to the tone and intent of headlines, not just their formatting. Industry experts have expressed concern. Bastian Grimm noted this shift from matching queries to reframing content for perceived performance. Brodie Clark highlighted instances where the meaning of articles was lost in the rewrites. Editors like Nilay Patel have criticized the changes, calling the new headlines the worst kind of slop, while James Ball questioned whether media publishers retain enough influence to push back against such alterations.

Separately, the March 2026 spam update concluded in a remarkably short timeframe. The rollout began on March 24 and was fully completed by the morning of March 25, lasting under 20 hours. This speed is unprecedented compared to recent updates, which have taken days or even weeks. The update applies globally to all languages, and Google did not announce any new spam policies alongside it. The quiet community response suggests a limited visible impact for most sites. SEO professionals like Nilesh Pansuriya have pointed out the historical speed of this update, noting it finished before many in the industry even realized it had started. For webmasters, the clear date range of March 24-25 is the key period to review in Search Console for any traffic fluctuations.

In another development, Google has updated its structured data documentation for Discussion Forum and Q&A pages. The new guidelines introduce a recommended property called `digitalSourceType`, which allows site owners to label content generated by AI models or automated bots. The property uses standard IPTC media codes to distinguish between content from a trained model and that from a simpler automated process. Because the property is only recommended and not required, Google will assume any unlabeled content is human-generated. Jan-Willem Bobbink identified this as a potential loophole, contrasting it with stricter requirements for product feeds, and questioning the effectiveness of a voluntary system for forums.

Meanwhile, Bing is offering publishers greater transparency. Its Webmaster Tools AI Performance dashboard now includes a feature that maps grounding queries to cited pages. This allows webmasters to see which specific pages on their site are being cited for particular AI queries in Copilot or Bing AI summaries, and vice versa. While the data is still a sample, this functionality helps publishers understand which content is winning visibility in AI experiences. Aleyda Solís welcomed the feature as a response to community feedback, and Navah Hopkins explained that grounding queries reveal how AI interprets user intent, with the page-level report showing which content is most effective.

A common theme emerges from these updates. Google is increasingly asserting control over content presentation, from rewriting headlines to rapidly enforcing spam policies and creating frameworks for content origin labeling. In contrast, Bing is focusing on providing publishers with more detailed data and visibility into how their content performs within AI-generated answers, addressing a measurement gap that currently exists in Google’s Search Console.