What happened
Google has introduced a significant shift in how breaking news is presented by embedding the "Top Stories" carousel directly within AI Overviews (formerly known as SGE). This update places publishers and digital marketers in a complex situation. Specifically, if a website utilizes the "opt-out" settings in Google Search Console to prevent Google's AI models from using their content, they risk being excluded from the news carousel whenever it appears as part of an AI-generated summary.
Technology context
At the heart of this issue is Google-Extended, a protocol that allows site owners to signal whether their data can be used to improve AI models like Gemini or Vertex AI. Initially, this was presented as a way to control AI training without affecting traditional search indexing. However, as AI Overviews increasingly become the primary interface for informational queries, the boundary between "AI training" and "search visibility" is blurring. The system now seems to treat consent for AI usage as a prerequisite for inclusion in high-visibility AI-driven search features.
Why it matters
For the media industry, the stakes are incredibly high. The "Top Stories" section is the lifeblood of organic traffic for news organizations. By linking this visibility to AI data permissions, Google is essentially forcing a trade-off: publishers must choose between protecting their intellectual property from generative models or maintaining their audience reach. This move has sparked debate regarding "digital coercion," where access to a platform's user base is leveraged to gain free training data for commercial AI products.
Key terms explained
- AI Overviews: AI-generated responses that appear at the top of Google Search results, providing a synthesized answer to user queries.
- Top Stories: A prominent search feature that displays a carousel of news articles related to trending or timely topics.
- Google-Extended: A user-agent token that gives webmasters the ability to control whether their site content is used to train Google's generative AI models.
- SERP (Search Engine Results Page): The page displayed by a search engine in response to a query by a user.
Impact
In the short term, we expect to see a traffic decline for publishers who have taken a hardline stance against AI data scraping. In the medium term, this could lead to a "forced compliance" scenario where websites prioritize visibility over copyright concerns to ensure economic survival. Furthermore, this integration may trigger new legal challenges as the distinction between fair use for search indexing and the commercial exploitation of data for AI training becomes harder to define.
What's next
Predictions indicate that Google will further integrate traditional search modules—such as shopping, local maps, and flights—into AI Overviews. Publishers will need to evolve their SEO strategies from keyword-centric approaches to "entity-based" optimization to remain visible within generative frameworks. We may also see increased regulatory scrutiny, particularly in regions like the EU, to determine if search engines can legally bundle search visibility with AI training consent.
Educational analysis generated with AI and editorially reviewed.
Sources: Search Engine Journal, Google Search Central Blog.