SEOIntel Weekly News Round-up (Second Week of October 2025)

This week in search has been a busy one, with big moves shaping both AI and regulation. Google is expanding AI Mode globally, bringing its conversational search experience to more countries and languages. At the same time, a new study sheds light on how AI citations actually work—showing that brand-controlled sources, location, and context all […]
Marie Aquino
October 10, 2025

This week in search has been a busy one, with big moves shaping both AI and regulation. Google is expanding AI Mode globally, bringing its conversational search experience to more countries and languages. At the same time, a new study sheds light on how AI citations actually work—showing that brand-controlled sources, location, and context all play a major role in what gets surfaced.

Across the pond, the UK’s Competition and Markets Authority (CMA) has officially activated its powers over Google Search, opening the door to stricter oversight.

And finally, Microsoft has stepped in with guidance for publishers and advertisers on how to optimize content for AI-driven search answers.

Google AI Mode Expands Globally: More Languages, More Reach

Google has announced a sweeping expansion of AI Mode in Search: it’s now launching in more than 35 new languages and rolling out across 40+ additional countries and territories. With this update, AI Mode is available in over 200 countries and territories in total, making its conversational, AI-powered search experience more accessible globally.

Under the hood, these capabilities are powered by Google’s custom Gemini 2.5 model, which enables AI Mode to handle multimodal inputs (text, image, voice) and deeper reasoning in local languages. The idea is not just translation, but making AI Mode locally relevant—so it understands cultural nuance, idioms, and language-specific contexts.

Already, Google is seeing users ask longer, more detailed queries in AI Mode—questions that might have taken multiple search steps before. As this rolls out, more people will be able to engage with AI-powered search in their preferred language, and explore topics more deeply without language barriers.

What SEOs & Site Owners Should Watch

  • Multilingual readiness: If your site serves global audiences, now is the time to ensure your content is well localized—beyond translation, considering cultural context and phrasing.
  • Structured and rich content matters more: AI Mode needs strong signals (structured data, clear headings, image context) to surface your site in non-English queries.
  • Attribution & analytics: As AI Mode expands, traffic patterns will evolve. Monitor how much visibility comes via AI Mode in other languages and how click behavior shifts.
  • Performance & localization: Page speed, language switching, hreflang, and server location may become even more important in these new markets.
  • Content quality & trust signals: Since AI answers in local languages will draw from your content, maintaining accuracy, authoritativeness, and verifying information is crucial.

The Bigger Picture

This global rollout shows how fast Google is moving to make AI Mode a default part of search worldwide. For businesses and SEOs, the challenge isn’t just ranking in English anymore—it’s making sure your brand is discoverable, trustworthy, and relevant across many languages and cultures. The search landscape is becoming more multilingual, multimodal, and AI-driven—those who adapt will find the biggest opportunities.


Study Reveals AI Citations Are Mostly Controlled by Brands — and Heavily Influenced by Location & Context

A new landmark study by Yext analyzed 6.8 million AI citations drawn from 1.6 million responses across major AI models (ChatGPT / OpenAI, Gemini, Perplexity). The core idea: to truly understand how AI search works, we need to move beyond a brand-level view and start factoring in user location, query context, and intent.

Key Findings in a Nutshell

  • Brand-controlled sources dominate citations. The study found that 86% of AI citations come from sources that brands already own or manage (websites, business listings, reviews) — not Reddit, forums, or random third-party sites.
  • Websites and listings lead the pack. First-party websites accounted for 44% of citations, while directory/listing sources made up another 42%. Other sources (reviews, social) contributed ~8%, and forums just ~2%.
  • AI model preferences vary. Gemini leaned more heavily on websites, OpenAI favored listings, and Perplexity pulled from a broader mix of sources including industry-specific directories.
  • Location and context shift citation behavior. When you control for user location and query intent (branded vs unbranded, objective vs subjective), citation patterns change—showing why brand-level studies often miss nuances.
  • Industry differences matter. In sectors like food service, reviews and reputation sources play a stronger role. In healthcare, listings (medical directories) are especially critical.

Why This Study Matters

This research challenges the popular myth that forums, memes, or fringe sites dominate AI answers. Instead, control matters: the sources you already manage play a heavy role in getting cited. That means brands haven’t lost their voice—they have an opportunity to optimize how their content appears within AI responses.

What You Should Do Now

  • Focus on first-party properties & listings. Since most citations come from sites you control, keep them authoritative, structured, and up-to-date.
  • Mind location and intent. Tailor content for local audiences and specific query types (objective vs subjective). The “global brand” approach alone won’t always work.
  • Use structured data, schema, and modular content. Break content into segments AI can cite individually.
  • Monitor models individually. OpenAI, Gemini, and Perplexity behave differently—know which sources each favors for your vertical.
  • Track citations, not just traffic. Traditional SEO metrics don’t capture how often your content is cited in AI answers.

The Yext study makes one thing clear: AI citations aren’t random—they’re structured, predictable, and often brand-controlled. For SEOs and marketers, that means opportunity is still on the table. By investing in your own digital properties, keeping listings accurate, and aligning content with user context, you’re not just chasing rankings—you’re positioning your brand as the trusted source AI models turn to. As AI search keeps evolving, the brands that adapt fastest will be the ones most frequently cited, and ultimately, the ones most visible.


Google Faces New UK Rules as CMA Activates Powers Over Search

The United Kingdom’s Competition and Markets Authority (CMA) has formally designated Google’s search and search advertising services with Strategic Market Status (SMS) under its new digital markets regime. This move is a landmark: it’s the first time a tech company in the UK has received this special status, which gives the regulator stronger powers to mandate changes in how Google runs search.

The designation doesn’t mean Google is being punished—it’s not a finding of wrongdoing—but rather that Google’s dominance (over 90% of UK search queries) makes it subject to ongoing oversight and behavioral or structural interventions.

The CMA has already laid out a roadmap for possible remedial measures, ranging from “choice screens” for search engines, fair ranking rules, to giving publishers more control over how their content is used in AI-generated summaries or results.

Why This Matters for Search Ecosystem

  1. More oversight & forced transparency
    Under SMS, Google will have to operate under stricter scrutiny. The CMA can require them to disclose or alter ranking logic, offer default choices to users, and prevent preferential treatment of Google’s own services or AI features.
  2. Impact on Google’s AI features & content usage
    Because AI Overviews and AI Mode are included in the scope of search services, the CMA can regulate how Google uses third-party content in generative summaries. That could mean giving publishers more say over usage, attribution, or opting out.
  3. Choice screens and search competition
    One proposed remedy is to force Google to show a “choice screen” to UK users, allowing them to pick alternative search engines (including AI assistants) at first use. This may open paths for rivals to gain visibility in a market dominated by Google.
  4. Pressure on publishers & content creators
    Publishers may gain leverage: for example, the rules might require Google to provide more transparency about how content is used and more control over AI-generated versions of their pages.
  5. Balance between regulation and innovation
    Google argues that heavy regulation could stifle innovation, especially in AI and new search features. The challenge ahead is whether the CMA’s interventions are sufficiently “proportionate and evidence-based” as Google and other stakeholders push back.

Timeline & Next Steps

  • The CMA began investigating Google under its new powers earlier in 2025, under the Digital Markets, Competition and Consumers Act 2024.
  • A public consultation on the proposed SMS designation closed in July 2025.
  • A final decision on designation is expected by October 13, 2025. Once designated, the CMA could begin imposing interventions (or “conduct requirements”) shortly afterward.

What SEOs & Publishers Should Watch

  • Keep track of intervention proposals — which rules will the CMA actually enforce? Some (like choice screens) are more likely; others (deep ranking changes) may be harder to implement.
  • Prepare content & usage contracts — understand how your content might be used in AI summaries and whether future rules could allow you to opt out or demand attribution.
  • Monitor traffic & rankings — changes to ranking or index rules could shift how pages are surfaced. Early detection will matter.
  • Be ready for new competitor entries — choice screens may let AI rivals or smaller search engines gain traction in the UK market.
  • Adjust expectations — if Google must abide by stricter rules, feature rollouts (especially AI features) in the UK could slow, or diverge from Google’s global versions.

Bottom Line

The CMA’s designation of Google Search with Strategic Market Status marks a turning point in how powerful digital platforms are regulated in the UK. While the exact remedies are still being shaped, the direction is clear: more transparency, stronger competition, and greater control for publishers and users alike. For SEOs and businesses, the landscape may soon look different—especially if choice screens, AI content rules, or ranking disclosures are enforced. Whether these changes spur healthier competition or add friction to innovation, one thing is certain: the UK is now a critical testing ground for how the future of search will be governed.


Microsoft Publishes Guide on Optimizing Content for AI Search Answers

Microsoft recently released a blog post titled “Optimizing Your Content for Inclusion in AI Search Answers” on October 8, 2025, outlining how content creators can better position their pages for selection in AI-driven response experiences.

The post begins by highlighting a dramatic rise in AI referrals: in June 2025 alone, sites saw a 357% year-over-year increase in AI-driven traffic, reaching 1.13 billion visits. As Microsoft sees it, the game has shifted: it’s no longer enough to just rank well, your content must also be selected to appear in AI assistant responses. AI systems don’t view entire pages—they parse them into fragments, evaluate them individually, and then assemble answers from multiple sources.

To help publishers, Microsoft offers practical guidance on how to make your content more AI-friendly. Titles, meta descriptions, and H1 tags should clearly reflect purpose and intent. Headings (H2, H3) need to break down content into logical, isolated ideas. Using Q&A blocks, bulleted lists, and tables increases the likelihood that AI will lift useful fragments directly. Schema markup also helps AI systems interpret types of content—FAQ, product, review—but it’s not a guarantee of inclusion.

Microsoft also warns against common mistakes that reduce AI search visibility:

  • Avoid long, unbroken paragraphs (they hinder parsing)
  • Don’t hide key answers behind tabs or collapsible content
  • Refrain from using PDFs or embedding essential info only in images (without text equivalents)

Still, there’s no silver bullet. Microsoft notes that no set of tactics can guarantee presentation in an AI answer—selection is competitive and contextual. Clear, modular, semantically strong content gives you the best odds.


As the AI-powered search landscape continues to expand and regulators tighten their grip, one theme is clear: adaptation is non-negotiable. Google’s global rollout of AI Mode means more users will start expecting conversational, context-aware results, while the Yext study shows us just how much brands can (and can’t) control their visibility in AI citations.

Meanwhile, the UK CMA’s designation reminds us that regulatory frameworks are catching up fast, and Microsoft’s new playbook highlights the growing need to align content with AI search. The shifts are coming quickly—and those who pivot early will be the ones best positioned to thrive.